fix tool parsing to load json if json

## Motivation

failed again for kimi on a machine that had never downloaded it.

## Test Plan

### Manual Testing
it worked this time

.mlx_typings/mlx/core/__init__.pyi

@@ -215,6 +215,22 @@ class StreamContext:
         traceback: object | None = ...,
     ) -> None: ...
 
+def device_info() -> dict[str, str | int]:
+    """
+    Get information about the GPU device and system settings.
+
+    Currently returns:
+
+    * ``architecture``
+    * ``max_buffer_size``
+    * ``max_recommended_working_set_size``
+    * ``memory_size``
+    * ``resource_limit``
+
+    Returns:
+        dict: A dictionary with string keys and string or integer values.
+    """
+
 def abs(a: array, /, *, stream: Stream | Device | None = ...) -> array:
     """
     Element-wise absolute value.

app/EXO/EXO/ContentView.swift

@@ -15,14 +15,23 @@ struct ContentView: View {
     @EnvironmentObject private var localNetworkChecker: LocalNetworkChecker
     @EnvironmentObject private var updater: SparkleUpdater
     @EnvironmentObject private var thunderboltBridgeService: ThunderboltBridgeService
+    @EnvironmentObject private var settingsWindowController: SettingsWindowController
     @State private var focusedNode: NodeViewModel?
     @State private var deletingInstanceIDs: Set<String> = []
     @State private var showAllNodes = false
     @State private var showAllInstances = false
+    @State private var baseURLCopied = false
     @State private var showAdvanced = false
     @State private var showDebugInfo = false
-    @State private var bugReportInFlight = false
-    @State private var bugReportMessage: String?
+    private enum BugReportPhase: Equatable {
+        case idle
+        case prompting
+        case sending(String)
+        case success(String)
+        case failure(String)
+    }
+    @State private var bugReportPhase: BugReportPhase = .idle
+    @State private var bugReportUserDescription: String = ""
     @State private var uninstallInProgress = false
     @State private var pendingNamespace: String = ""
     @State private var pendingHFToken: String = ""
@@ -258,139 +267,79 @@ struct ContentView: View {
         VStack(alignment: .leading, spacing: 0) {
             if controller.status != .stopped {
                 dashboardButton
+                baseURLRow
                 Divider()
                     .padding(.vertical, 8)
             } else {
                 Divider()
                     .padding(.vertical, 4)
             }
-            advancedSection
-                .padding(.bottom, 8)
-            controlButton(title: "Quit", tint: .secondary) {
+            HoverButton(
+                title: "Settings",
+                tint: .primary,
+                trailingSystemImage: "gear"
+            ) {
+                settingsWindowController.open(
+                    controller: controller,
+                    updater: updater,
+                    networkStatusService: networkStatusService,
+                    thunderboltBridgeService: thunderboltBridgeService,
+                    stateService: stateService
+                )
+            }
+            HoverButton(
+                title: "Check for Updates",
+                tint: .primary,
+                trailingSystemImage: "arrow.triangle.2.circlepath"
+            ) {
+                updater.checkForUpdates()
+            }
+            .padding(.bottom, 8)
+            HoverButton(title: "Quit", tint: .secondary) {
                 controller.stop()
                 NSApplication.shared.terminate(nil)
             }
         }
     }
 
-    private var advancedSection: some View {
-        VStack(alignment: .leading, spacing: 6) {
-            HStack {
-                Text("Advanced")
-                    .font(.caption)
-                    .foregroundColor(.secondary)
-                Spacer()
-                collapseButton(isExpanded: $showAdvanced)
-            }
-            .animation(nil, value: showAdvanced)
-            if showAdvanced {
-                VStack(alignment: .leading, spacing: 8) {
-                    VStack(alignment: .leading, spacing: 4) {
-                        Text("Cluster Namespace")
-                            .font(.caption2)
-                            .foregroundColor(.secondary)
-                        HStack {
-                            TextField("optional", text: $pendingNamespace)
-                                .textFieldStyle(.roundedBorder)
-                                .font(.caption2)
-                                .onAppear {
-                                    pendingNamespace = controller.customNamespace
-                                }
-                            Button("Save & Restart") {
-                                controller.customNamespace = pendingNamespace
-                                if controller.status == .running || controller.status == .starting {
-                                    controller.restart()
-                                }
-                            }
-                            .font(.caption2)
-                            .disabled(pendingNamespace == controller.customNamespace)
-                        }
-                    }
-                    VStack(alignment: .leading, spacing: 4) {
-                        Text("HuggingFace Token")
-                            .font(.caption2)
-                            .foregroundColor(.secondary)
-                        HStack {
-                            SecureField("optional", text: $pendingHFToken)
-                                .textFieldStyle(.roundedBorder)
-                                .font(.caption2)
-                                .onAppear {
-                                    pendingHFToken = controller.hfToken
-                                }
-                            Button("Save & Restart") {
-                                controller.hfToken = pendingHFToken
-                                if controller.status == .running || controller.status == .starting {
-                                    controller.restart()
-                                }
-                            }
-                            .font(.caption2)
-                            .disabled(pendingHFToken == controller.hfToken)
-                        }
-                    }
-                    Divider()
-                    HStack {
-                        Toggle(
-                            "Enable Image Models (experimental)", isOn: $pendingEnableImageModels
-                        )
-                        .toggleStyle(.switch)
-                        .font(.caption2)
-                        .onAppear {
-                            pendingEnableImageModels = controller.enableImageModels
-                        }
-
-                        Spacer()
-
-                        Button("Save & Restart") {
-                            controller.enableImageModels = pendingEnableImageModels
-                            if controller.status == .running || controller.status == .starting {
-                                controller.restart()
-                            }
-                        }
-                        .font(.caption2)
-                        .disabled(pendingEnableImageModels == controller.enableImageModels)
-                    }
-                    HoverButton(title: "Check for Updates", small: true) {
-                        updater.checkForUpdates()
-                    }
-                    debugSection
-                    HoverButton(title: "Uninstall", tint: .red, small: true) {
-                        showUninstallConfirmationAlert()
-                    }
-                    .disabled(uninstallInProgress)
-                }
-                .transition(.opacity)
-            }
-        }
-        .animation(.easeInOut(duration: 0.25), value: showAdvanced)
-    }
-
-    private func controlButton(title: String, tint: Color = .primary, action: @escaping () -> Void)
-        -> some View
-    {
-        HoverButton(title: title, tint: tint, trailingSystemImage: nil, action: action)
-    }
-
     private var dashboardButton: some View {
-        Button {
+        HoverButton(
+            title: "Web Dashboard",
+            tint: .primary,
+            trailingSystemImage: "arrow.up.right"
+        ) {
             guard let url = URL(string: "http://localhost:52415/") else { return }
             NSWorkspace.shared.open(url)
-        } label: {
-            HStack {
-                Image(systemName: "arrow.up.right.square")
-                    .imageScale(.small)
-                Text("Dashboard")
-                    .fontWeight(.medium)
-                Spacer()
-            }
-            .padding(.vertical, 8)
-            .padding(.horizontal, 10)
-            .background(
-                RoundedRectangle(cornerRadius: 8, style: .continuous)
-                    .fill(Color(red: 1.0, green: 0.87, blue: 0.0).opacity(0.2))
-            )
         }
-        .buttonStyle(.plain)
-        .padding(.bottom, 4)
+    }
+
+    private var baseURLRow: some View {
+        HStack(spacing: 6) {
+            Image(systemName: "link")
+                .imageScale(.small)
+                .foregroundColor(.secondary)
+            Text("localhost:52415/v1")
+                .font(.system(.caption, design: .monospaced))
+                .foregroundColor(.primary)
+            Spacer()
+            Button {
+                NSPasteboard.general.clearContents()
+                NSPasteboard.general.setString("http://localhost:52415/v1", forType: .string)
+                baseURLCopied = true
+                DispatchQueue.main.asyncAfter(deadline: .now() + 2) {
+                    baseURLCopied = false
+                }
+            } label: {
+                Image(systemName: baseURLCopied ? "checkmark" : "doc.on.doc")
+                    .imageScale(.small)
+                    .foregroundColor(baseURLCopied ? .green : .secondary)
+                    .contentTransition(.symbolEffect(.replace))
+            }
+            .buttonStyle(.plain)
+            .help("Copy API base URL")
+        }
+        .padding(.vertical, 4)
+        .padding(.horizontal, 8)
     }
 
     private func collapseButton(isExpanded: Binding<Bool>) -> some View {
@@ -611,39 +560,115 @@ struct ContentView: View {
     }
 
     private var sendBugReportButton: some View {
-        VStack(alignment: .leading, spacing: 4) {
-            Button {
-                Task {
-                    await sendBugReport()
-                }
-            } label: {
-                HStack {
-                    if bugReportInFlight {
-                        ProgressView()
-                            .scaleEffect(0.6)
+        VStack(alignment: .leading, spacing: 6) {
+            switch bugReportPhase {
+            case .idle:
+                Button {
+                    bugReportPhase = .prompting
+                    bugReportUserDescription = ""
+                } label: {
+                    HStack {
+                        Text("Send Bug Report")
+                            .font(.caption)
+                            .fontWeight(.semibold)
+                        Spacer()
                     }
-                    Text("Send Bug Report")
-                        .font(.caption)
-                        .fontWeight(.semibold)
-                    Spacer()
+                    .padding(.vertical, 6)
+                    .padding(.horizontal, 8)
+                    .background(
+                        RoundedRectangle(cornerRadius: 6)
+                            .fill(Color.accentColor.opacity(0.12))
+                    )
                 }
-                .padding(.vertical, 6)
-                .padding(.horizontal, 8)
+                .buttonStyle(.plain)
+
+            case .prompting:
+                VStack(alignment: .leading, spacing: 6) {
+                    Text("What's the issue? (optional)")
+                        .font(.caption2)
+                        .foregroundColor(.secondary)
+                    TextEditor(text: $bugReportUserDescription)
+                        .font(.caption2)
+                        .frame(height: 60)
+                        .overlay(
+                            RoundedRectangle(cornerRadius: 4)
+                                .stroke(Color.secondary.opacity(0.3), lineWidth: 1)
+                        )
+                    HStack(spacing: 8) {
+                        Button("Send") {
+                            Task {
+                                await sendBugReport()
+                            }
+                        }
+                        .font(.caption2)
+                        .buttonStyle(.borderedProminent)
+                        .controlSize(.small)
+                        Button("Cancel") {
+                            bugReportPhase = .idle
+                        }
+                        .font(.caption2)
+                        .buttonStyle(.bordered)
+                        .controlSize(.small)
+                    }
+                }
+                .padding(8)
                 .background(
                     RoundedRectangle(cornerRadius: 6)
-                        .fill(Color.accentColor.opacity(0.12))
+                        .fill(Color.accentColor.opacity(0.06))
                 )
-            }
-            .buttonStyle(.plain)
-            .disabled(bugReportInFlight)
 
-            if let message = bugReportMessage {
-                Text(message)
+            case .sending(let message):
+                HStack(spacing: 6) {
+                    ProgressView()
+                        .scaleEffect(0.6)
+                    Text(message)
+                        .font(.caption2)
+                        .foregroundColor(.secondary)
+                }
+
+            case .success(let message):
+                VStack(alignment: .leading, spacing: 6) {
+                    Text(message)
+                        .font(.caption2)
+                        .foregroundColor(.secondary)
+                        .fixedSize(horizontal: false, vertical: true)
+                    Button {
+                        openGitHubIssue()
+                    } label: {
+                        HStack(spacing: 4) {
+                            Image(systemName: "arrow.up.right.square")
+                                .imageScale(.small)
+                            Text("Create GitHub Issue")
+                                .font(.caption2)
+                        }
+                    }
+                    .buttonStyle(.bordered)
+                    .controlSize(.small)
+                    Button("Done") {
+                        bugReportPhase = .idle
+                        bugReportUserDescription = ""
+                    }
                     .font(.caption2)
+                    .buttonStyle(.plain)
                     .foregroundColor(.secondary)
-                    .fixedSize(horizontal: false, vertical: true)
+                }
+
+            case .failure(let message):
+                VStack(alignment: .leading, spacing: 4) {
+                    Text(message)
+                        .font(.caption2)
+                        .foregroundColor(.red)
+                        .fixedSize(horizontal: false, vertical: true)
+                    Button("Dismiss") {
+                        bugReportPhase = .idle
+                    }
+                    .font(.caption2)
+                    .buttonStyle(.plain)
+                    .foregroundColor(.secondary)
+                }
             }
         }
+        .animation(.easeInOut(duration: 0.2), value: bugReportPhase)
     }
 
     private var processToggleBinding: Binding<Bool> {
@@ -687,16 +712,58 @@ struct ContentView: View {
     }
 
     private func sendBugReport() async {
-        bugReportInFlight = true
-        bugReportMessage = "Collecting logs..."
+        bugReportPhase = .sending("Collecting logs...")
         let service = BugReportService()
+        let description = bugReportUserDescription.trimmingCharacters(in: .whitespacesAndNewlines)
         do {
-            let outcome = try await service.sendReport(isManual: true)
-            bugReportMessage = outcome.message
+            let outcome = try await service.sendReport(
+                isManual: true,
+                userDescription: description.isEmpty ? nil : description
+            )
+            if outcome.success {
+                bugReportPhase = .success(outcome.message)
+            } else {
+                bugReportPhase = .failure(outcome.message)
+            }
         } catch {
-            bugReportMessage = error.localizedDescription
+            bugReportPhase = .failure(error.localizedDescription)
+        }
+    }
+
+    private func openGitHubIssue() {
+        let description = bugReportUserDescription.trimmingCharacters(in: .whitespacesAndNewlines)
+
+        var bodyParts: [String] = []
+        bodyParts.append("## Describe the bug")
+        bodyParts.append("")
+        if !description.isEmpty {
+            bodyParts.append(description)
+        } else {
+            bodyParts.append("A clear and concise description of what the bug is.")
+        }
+        bodyParts.append("")
+        bodyParts.append("## Environment")
+        bodyParts.append("")
+        bodyParts.append("- macOS Version: \(ProcessInfo.processInfo.operatingSystemVersionString)")
+        bodyParts.append("- EXO Version: \(buildTag) (\(buildCommit))")
+        bodyParts.append("")
+        bodyParts.append("## Additional context")
+        bodyParts.append("")
+        bodyParts.append("A bug report with diagnostic logs was submitted via the app.")
+
+        let body = bodyParts.joined(separator: "\n")
+
+        var components = URLComponents(string: "https://github.com/exo-explore/exo/issues/new")!
+        components.queryItems = [
+            URLQueryItem(name: "template", value: "bug_report.md"),
+            URLQueryItem(name: "title", value: "[BUG] "),
+            URLQueryItem(name: "body", value: body),
+            URLQueryItem(name: "labels", value: "bug"),
+        ]
+
+        if let url = components.url {
+            NSWorkspace.shared.open(url)
         }
-        bugReportInFlight = false
     }
 
     private func showUninstallConfirmationAlert() {

app/EXO/EXO/EXOApp.swift

@@ -21,7 +21,9 @@ struct EXOApp: App {
     @StateObject private var localNetworkChecker: LocalNetworkChecker
     @StateObject private var updater: SparkleUpdater
     @StateObject private var thunderboltBridgeService: ThunderboltBridgeService
+    @StateObject private var settingsWindowController: SettingsWindowController
     private let terminationObserver: TerminationObserver
+    private let firstLaunchPopout = FirstLaunchPopout()
     private let ciContext = CIContext(options: nil)
 
     init() {
@@ -43,12 +45,13 @@ struct EXOApp: App {
         _updater = StateObject(wrappedValue: updater)
         let thunderboltBridge = ThunderboltBridgeService(clusterStateService: service)
         _thunderboltBridgeService = StateObject(wrappedValue: thunderboltBridge)
+        _settingsWindowController = StateObject(wrappedValue: SettingsWindowController())
         enableLaunchAtLoginIfNeeded()
         // Install LaunchDaemon to disable Thunderbolt Bridge on startup (prevents network loops)
         NetworkSetupHelper.promptAndInstallIfNeeded()
         // Check local network access periodically (warning disappears when user grants permission)
         localNetwork.startPeriodicChecking(interval: 10)
-        controller.scheduleLaunch(after: 15)
+        controller.scheduleLaunch(after: 5)
         service.startPolling()
         networkStatus.startPolling()
     }
@@ -62,8 +65,19 @@ struct EXOApp: App {
                 .environmentObject(localNetworkChecker)
                 .environmentObject(updater)
                 .environmentObject(thunderboltBridgeService)
+                .environmentObject(settingsWindowController)
         } label: {
             menuBarIcon
+                .onReceive(controller.$isFirstLaunchReady) { ready in
+                    if ready {
+                        DispatchQueue.main.asyncAfter(deadline: .now() + 3.0) {
+                            self.firstLaunchPopout.onComplete = { [weak controller] in
+                                controller?.markOnboardingCompleted()
+                            }
+                            self.firstLaunchPopout.show()
+                        }
+                    }
+                }
         }
         .menuBarExtraStyle(.window)
     }

app/EXO/EXO/ExoProcessController.swift

@@ -5,6 +5,7 @@ import Foundation
 private let customNamespaceKey = "EXOCustomNamespace"
 private let hfTokenKey = "EXOHFToken"
 private let enableImageModelsKey = "EXOEnableImageModels"
+private let onboardingCompletedKey = "EXOOnboardingCompleted"
 
 @MainActor
 final class ExoProcessController: ObservableObject {
@@ -60,6 +61,9 @@ final class ExoProcessController: ObservableObject {
         }
     }
 
+    /// Fires once when EXO transitions to `.running` for the very first time (fresh install).
+    @Published private(set) var isFirstLaunchReady = false
+
     private var process: Process?
     private var runtimeDirectoryURL: URL?
     private var pendingLaunchTask: Task<Void, Never>?
@@ -113,6 +117,9 @@ final class ExoProcessController: ObservableObject {
             try child.run()
             process = child
             status = .running
+
+            // Show welcome popout on every launch
+            isFirstLaunchReady = true
         } catch {
             process = nil
             status = .failed(message: "Launch error")
@@ -164,6 +171,17 @@ final class ExoProcessController: ObservableObject {
         launch()
     }
 
+    /// Mark onboarding as completed (user interacted with the welcome popout).
+    func markOnboardingCompleted() {
+        UserDefaults.standard.set(true, forKey: onboardingCompletedKey)
+    }
+
+    /// Reset onboarding so the welcome popout appears on next launch.
+    func resetOnboarding() {
+        UserDefaults.standard.removeObject(forKey: onboardingCompletedKey)
+        isFirstLaunchReady = false
+    }
+
     func scheduleLaunch(after seconds: TimeInterval) {
         cancelPendingLaunch()
         let start = max(1, Int(ceil(seconds)))

app/EXO/EXO/Services/BugReportService.swift

@@ -38,7 +38,8 @@ struct BugReportService {
     func sendReport(
         baseURL: URL = URL(string: "http://127.0.0.1:52415")!,
         now: Date = Date(),
-        isManual: Bool = false
+        isManual: Bool = false,
+        userDescription: String? = nil
     ) async throws -> BugReportOutcome {
         let timestamp = Self.runTimestampString(now)
         let dayPrefix = Self.dayPrefixString(now)
@@ -60,7 +61,8 @@ struct BugReportService {
             ifconfig: ifconfigText,
             debugInfo: debugInfo,
             isManual: isManual,
-            clusterTbBridgeStatus: clusterTbBridgeStatus
+            clusterTbBridgeStatus: clusterTbBridgeStatus,
+            userDescription: userDescription
         )
 
         let eventLogFiles = readAllEventLogs()
@@ -306,7 +308,8 @@ struct BugReportService {
         ifconfig: String,
         debugInfo: DebugInfo,
         isManual: Bool,
-        clusterTbBridgeStatus: [[String: Any]]?
+        clusterTbBridgeStatus: [[String: Any]]?,
+        userDescription: String? = nil
     ) -> Data? {
         let system = readSystemMetadata()
         let exo = readExoMetadata()
@@ -323,6 +326,9 @@ struct BugReportService {
         if let tbStatus = clusterTbBridgeStatus {
             payload["cluster_thunderbolt_bridge"] = tbStatus
         }
+        if let desc = userDescription, !desc.isEmpty {
+            payload["user_description"] = desc
+        }
         return try? JSONSerialization.data(withJSONObject: payload, options: [.prettyPrinted])
     }
 

app/EXO/EXO/Views/FirstLaunchPopout.swift

@@ -0,0 +1,200 @@
+import AppKit
+import SwiftUI
+
+/// A popover callout anchored to the menu bar icon on every launch,
+/// pointing the user to the web dashboard with an arrow connecting to the icon.
+@MainActor
+final class FirstLaunchPopout {
+    private var popover: NSPopover?
+    private var countdownTask: Task<Void, Never>?
+    private static let dashboardURL = "http://localhost:52415/"
+
+    /// Called when the user completes onboarding (clicks Open Dashboard or dismisses).
+    var onComplete: (() -> Void)?
+
+    func show() {
+        guard popover == nil else { return }
+
+        // The status bar button may not exist yet on first launch; retry generously.
+        showWithRetry(attemptsRemaining: 15)
+    }
+
+    private func showWithRetry(attemptsRemaining: Int) {
+        guard attemptsRemaining > 0 else {
+            // Exhausted retries — fall back to just opening the dashboard directly.
+            openDashboard()
+            onComplete?()
+            return
+        }
+
+        guard let button = Self.findStatusItemButton() else {
+            DispatchQueue.main.asyncAfter(deadline: .now() + 1.0) { [weak self] in
+                self?.showWithRetry(attemptsRemaining: attemptsRemaining - 1)
+            }
+            return
+        }
+
+        let pop = NSPopover()
+        pop.behavior = .applicationDefined
+        pop.animates = true
+        pop.contentSize = NSSize(width: 280, height: 120)
+        pop.contentViewController = NSHostingController(
+            rootView: WelcomeCalloutView(
+                countdownDuration: 10,
+                onDismiss: { [weak self] in
+                    self?.onComplete?()
+                    self?.dismiss()
+                },
+                onOpen: { [weak self] in
+                    self?.openDashboard()
+                    self?.onComplete?()
+                    self?.dismiss()
+                }
+            )
+        )
+
+        self.popover = pop
+        pop.show(relativeTo: button.bounds, of: button, preferredEdge: .minY)
+
+        // Auto-open dashboard after 10s then dismiss
+        countdownTask = Task {
+            try? await Task.sleep(nanoseconds: 10_000_000_000)
+            if !Task.isCancelled {
+                openDashboard()
+                onComplete?()
+                dismiss()
+            }
+        }
+    }
+
+    func dismiss() {
+        countdownTask?.cancel()
+        countdownTask = nil
+        guard let pop = popover else { return }
+        popover = nil
+        pop.performClose(nil)
+    }
+
+    private func openDashboard() {
+        guard let url = URL(string: Self.dashboardURL) else { return }
+        NSWorkspace.shared.open(url)
+    }
+
+    /// Finds the NSStatusBarButton created by SwiftUI's MenuBarExtra.
+    /// Walks the view hierarchy to find the actual button rather than the content view.
+    private static func findStatusItemButton() -> NSView? {
+        for window in NSApp.windows {
+            let className = NSStringFromClass(type(of: window))
+            // Match NSStatusBarWindow or any internal SwiftUI status bar window
+            guard className.contains("StatusBar") || className.contains("MenuBarExtra") else {
+                continue
+            }
+            if let content = window.contentView {
+                if let button = findButton(in: content) {
+                    return button
+                }
+                // Fall back to the content view itself if it has a non-zero frame
+                if content.frame.width > 0 {
+                    return content
+                }
+            }
+        }
+        return nil
+    }
+
+    /// Recursively searches the view hierarchy for an NSStatusBarButton.
+    private static func findButton(in view: NSView) -> NSView? {
+        let className = NSStringFromClass(type(of: view))
+        if className.contains("StatusBarButton") || className.contains("StatusItem") {
+            return view
+        }
+        for subview in view.subviews {
+            if let found = findButton(in: subview) {
+                return found
+            }
+        }
+        return nil
+    }
+}
+
+/// Minimal welcome callout — friendly pointer, not a wall of text.
+/// Rendered inside the NSPopover which provides its own chrome and arrow.
+private struct WelcomeCalloutView: View {
+    let countdownDuration: Int
+    let onDismiss: () -> Void
+    let onOpen: () -> Void
+    @State private var countdown: Int
+    @State private var timerTask: Task<Void, Never>?
+
+    init(countdownDuration: Int, onDismiss: @escaping () -> Void, onOpen: @escaping () -> Void) {
+        self.countdownDuration = countdownDuration
+        self.onDismiss = onDismiss
+        self.onOpen = onOpen
+        self._countdown = State(initialValue: countdownDuration)
+    }
+
+    var body: some View {
+        VStack(alignment: .leading, spacing: 10) {
+            HStack(alignment: .top) {
+                Text("EXO is running")
+                    .font(.system(.headline, design: .rounded))
+                    .fontWeight(.semibold)
+                    .foregroundColor(.primary)
+                Spacer()
+                Button {
+                    onDismiss()
+                } label: {
+                    Image(systemName: "xmark.circle.fill")
+                        .font(.system(size: 14))
+                        .foregroundStyle(.tertiary)
+                }
+                .buttonStyle(.plain)
+            }
+
+            Text("Run your first model here:")
+                .font(.system(.subheadline, design: .default))
+                .foregroundColor(.secondary)
+
+            HStack {
+                Button {
+                    onOpen()
+                } label: {
+                    Label("Open Dashboard", systemImage: "arrow.up.right.square")
+                        .font(.system(.caption, design: .default))
+                        .fontWeight(.medium)
+                }
+                .buttonStyle(.borderedProminent)
+                .tint(.accentColor)
+                .controlSize(.small)
+
+                Spacer()
+
+                if countdown > 0 {
+                    Text("Launching in \(countdown) secs...")
+                        .font(.system(.caption2, design: .default))
+                        .foregroundColor(.secondary.opacity(0.6))
+                        .monospacedDigit()
+                }
+            }
+        }
+        .padding(14)
+        .onAppear {
+            startCountdown()
+        }
+        .onDisappear {
+            timerTask?.cancel()
+            timerTask = nil
+        }
+    }
+
+    private func startCountdown() {
+        timerTask = Task {
+            while countdown > 0 {
+                try? await Task.sleep(nanoseconds: 1_000_000_000)
+                if !Task.isCancelled {
+                    countdown -= 1
+                }
+            }
+        }
+    }
+}

app/EXO/EXO/Views/SettingsView.swift

@@ -0,0 +1,478 @@
+import AppKit
+import SwiftUI
+
+/// Native macOS Settings window following Apple HIG.
+/// Organized into General, Model, Advanced, and About sections.
+struct SettingsView: View {
+    @EnvironmentObject private var controller: ExoProcessController
+    @EnvironmentObject private var updater: SparkleUpdater
+    @EnvironmentObject private var networkStatusService: NetworkStatusService
+    @EnvironmentObject private var thunderboltBridgeService: ThunderboltBridgeService
+    @EnvironmentObject private var stateService: ClusterStateService
+
+    @State private var pendingNamespace: String = ""
+    @State private var pendingHFToken: String = ""
+    @State private var pendingEnableImageModels = false
+    @State private var needsRestart = false
+    @State private var bugReportInFlight = false
+    @State private var bugReportMessage: String?
+    @State private var uninstallInProgress = false
+
+    var body: some View {
+        TabView {
+            generalTab
+                .tabItem {
+                    Label("General", systemImage: "gear")
+                }
+            modelTab
+                .tabItem {
+                    Label("Model", systemImage: "cube")
+                }
+            advancedTab
+                .tabItem {
+                    Label("Advanced", systemImage: "wrench.and.screwdriver")
+                }
+            aboutTab
+                .tabItem {
+                    Label("About", systemImage: "info.circle")
+                }
+        }
+        .frame(width: 450, height: 400)
+        .onAppear {
+            pendingNamespace = controller.customNamespace
+            pendingHFToken = controller.hfToken
+            pendingEnableImageModels = controller.enableImageModels
+            needsRestart = false
+        }
+    }
+
+    // MARK: - General Tab
+
+    private var generalTab: some View {
+        Form {
+            Section {
+                LabeledContent("Cluster Namespace") {
+                    TextField("default", text: $pendingNamespace)
+                        .textFieldStyle(.roundedBorder)
+                        .frame(width: 200)
+                }
+                Text("Nodes with the same namespace form a cluster. Leave empty for default.")
+                    .font(.caption)
+                    .foregroundColor(.secondary)
+            }
+
+            Section {
+                LabeledContent("HuggingFace Token") {
+                    SecureField("optional", text: $pendingHFToken)
+                        .textFieldStyle(.roundedBorder)
+                        .frame(width: 200)
+                }
+                Text("Required for gated models. Get yours at huggingface.co/settings/tokens")
+                    .font(.caption)
+                    .foregroundColor(.secondary)
+            }
+
+            Section {
+                HStack {
+                    Spacer()
+                    Button("Save & Restart") {
+                        applyGeneralSettings()
+                    }
+                    .disabled(!hasGeneralChanges)
+                }
+            }
+        }
+        .formStyle(.grouped)
+        .padding()
+    }
+
+    // MARK: - Model Tab
+
+    private var modelTab: some View {
+        Form {
+            Section {
+                Toggle("Enable Image Models (experimental)", isOn: $pendingEnableImageModels)
+                Text("Allow text-to-image and image-to-image models in the model picker.")
+                    .font(.caption)
+                    .foregroundColor(.secondary)
+            }
+
+            Section {
+                HStack {
+                    Spacer()
+                    Button("Save & Restart") {
+                        applyModelSettings()
+                    }
+                    .disabled(!hasModelChanges)
+                }
+            }
+        }
+        .formStyle(.grouped)
+        .padding()
+    }
+
+    // MARK: - Advanced Tab
+
+    private var advancedTab: some View {
+        Form {
+            Section("Onboarding") {
+                HStack {
+                    VStack(alignment: .leading) {
+                        Text("Reset Onboarding")
+                        Text("Opens the dashboard and resets the onboarding wizard.")
+                            .font(.caption)
+                            .foregroundColor(.secondary)
+                    }
+                    Spacer()
+                    Button("Reset") {
+                        guard let url = URL(string: "http://localhost:52415/?reset-onboarding")
+                        else { return }
+                        NSWorkspace.shared.open(url)
+                    }
+                }
+            }
+
+            Section("Debug Info") {
+                LabeledContent("Thunderbolt Bridge") {
+                    Text(thunderboltStatusText)
+                        .foregroundColor(thunderboltStatusColor)
+                }
+
+                VStack(alignment: .leading, spacing: 2) {
+                    clusterThunderboltBridgeView
+                }
+
+                VStack(alignment: .leading, spacing: 2) {
+                    interfaceIpList
+                }
+
+                VStack(alignment: .leading, spacing: 2) {
+                    rdmaStatusView
+                }
+
+                sendBugReportButton
+            }
+
+            Section("Danger Zone") {
+                Button(role: .destructive) {
+                    showUninstallConfirmationAlert()
+                } label: {
+                    HStack {
+                        Text("Uninstall EXO")
+                        Spacer()
+                        Image(systemName: "trash")
+                            .imageScale(.small)
+                    }
+                }
+                .disabled(uninstallInProgress)
+            }
+        }
+        .formStyle(.grouped)
+        .padding()
+    }
+
+    // MARK: - About Tab
+
+    private var aboutTab: some View {
+        Form {
+            Section {
+                LabeledContent("Version") {
+                    Text(buildTag)
+                        .textSelection(.enabled)
+                }
+                LabeledContent("Commit") {
+                    Text(buildCommit)
+                        .font(.system(.body, design: .monospaced))
+                        .textSelection(.enabled)
+                }
+            }
+
+            Section {
+                Button("Check for Updates") {
+                    updater.checkForUpdates()
+                }
+            }
+        }
+        .formStyle(.grouped)
+        .padding()
+    }
+
+    // MARK: - Debug Info Views (moved from ContentView)
+
+    private var thunderboltStatusText: String {
+        switch networkStatusService.status.thunderboltBridgeState {
+        case .some(.disabled):
+            return "Disabled"
+        case .some(.deleted):
+            return "Deleted"
+        case .some(.enabled):
+            return "Enabled"
+        case nil:
+            return "Unknown"
+        }
+    }
+
+    private var thunderboltStatusColor: Color {
+        switch networkStatusService.status.thunderboltBridgeState {
+        case .some(.disabled), .some(.deleted):
+            return .green
+        case .some(.enabled):
+            return .red
+        case nil:
+            return .secondary
+        }
+    }
+
+    private var clusterThunderboltBridgeView: some View {
+        let bridgeStatuses = stateService.latestSnapshot?.nodeThunderboltBridge ?? [:]
+        let localNodeId = stateService.localNodeId
+        let nodeProfiles = stateService.latestSnapshot?.nodeProfiles ?? [:]
+
+        return VStack(alignment: .leading, spacing: 1) {
+            if bridgeStatuses.isEmpty {
+                Text("Cluster TB Bridge: No data")
+                    .font(.caption2)
+                    .foregroundColor(.secondary)
+            } else {
+                Text("Cluster TB Bridge Status:")
+                    .font(.caption2)
+                    .foregroundColor(.secondary)
+                ForEach(Array(bridgeStatuses.keys.sorted()), id: \.self) { nodeId in
+                    if let status = bridgeStatuses[nodeId] {
+                        let nodeName =
+                            nodeProfiles[nodeId]?.friendlyName ?? String(nodeId.prefix(8))
+                        let isLocal = nodeId == localNodeId
+                        let prefix = isLocal ? "  \(nodeName) (local):" : "  \(nodeName):"
+                        let statusText =
+                            !status.exists
+                            ? "N/A"
+                            : (status.enabled ? "Enabled" : "Disabled")
+                        let color: Color =
+                            !status.exists
+                            ? .secondary
+                            : (status.enabled ? .red : .green)
+                        Text("\(prefix) \(statusText)")
+                            .font(.caption2)
+                            .foregroundColor(color)
+                    }
+                }
+            }
+        }
+    }
+
+    private var interfaceIpList: some View {
+        let statuses = networkStatusService.status.interfaceStatuses
+        return VStack(alignment: .leading, spacing: 1) {
+            Text("Interfaces (en0–en7):")
+                .font(.caption2)
+                .foregroundColor(.secondary)
+            if statuses.isEmpty {
+                Text("  Unknown")
+                    .font(.caption2)
+                    .foregroundColor(.secondary)
+            } else {
+                ForEach(statuses, id: \.interfaceName) { status in
+                    let ipText = status.ipAddress ?? "No IP"
+                    Text("  \(status.interfaceName): \(ipText)")
+                        .font(.caption2)
+                        .foregroundColor(status.ipAddress == nil ? .red : .green)
+                }
+            }
+        }
+    }
+
+    private var rdmaStatusView: some View {
+        let rdmaStatuses = stateService.latestSnapshot?.nodeRdmaCtl ?? [:]
+        let localNodeId = stateService.localNodeId
+        let nodeProfiles = stateService.latestSnapshot?.nodeProfiles ?? [:]
+        let localDevices = networkStatusService.status.localRdmaDevices
+        let localPorts = networkStatusService.status.localRdmaActivePorts
+
+        return VStack(alignment: .leading, spacing: 1) {
+            if rdmaStatuses.isEmpty {
+                Text("Cluster RDMA: No data")
+                    .font(.caption2)
+                    .foregroundColor(.secondary)
+            } else {
+                Text("Cluster RDMA Status:")
+                    .font(.caption2)
+                    .foregroundColor(.secondary)
+                ForEach(Array(rdmaStatuses.keys.sorted()), id: \.self) { nodeId in
+                    if let status = rdmaStatuses[nodeId] {
+                        let nodeName =
+                            nodeProfiles[nodeId]?.friendlyName ?? String(nodeId.prefix(8))
+                        let isLocal = nodeId == localNodeId
+                        let prefix = isLocal ? "  \(nodeName) (local):" : "  \(nodeName):"
+                        let statusText = status.enabled ? "Enabled" : "Disabled"
+                        let color: Color = status.enabled ? .green : .orange
+                        Text("\(prefix) \(statusText)")
+                            .font(.caption2)
+                            .foregroundColor(color)
+                    }
+                }
+            }
+            if !localDevices.isEmpty {
+                Text("  Local Devices: \(localDevices.joined(separator: ", "))")
+                    .font(.caption2)
+                    .foregroundColor(.secondary)
+            }
+            if !localPorts.isEmpty {
+                Text("  Local Active Ports:")
+                    .font(.caption2)
+                    .foregroundColor(.secondary)
+                ForEach(localPorts, id: \.device) { port in
+                    Text("    \(port.device) port \(port.port): \(port.state)")
+                        .font(.caption2)
+                        .foregroundColor(.green)
+                }
+            }
+        }
+    }
+
+    private var sendBugReportButton: some View {
+        VStack(alignment: .leading, spacing: 4) {
+            Button {
+                Task {
+                    await sendBugReport()
+                }
+            } label: {
+                HStack {
+                    if bugReportInFlight {
+                        ProgressView()
+                            .scaleEffect(0.6)
+                    }
+                    Text("Send Bug Report")
+                        .font(.caption)
+                        .fontWeight(.semibold)
+                    Spacer()
+                }
+            }
+            .disabled(bugReportInFlight)
+
+            if let message = bugReportMessage {
+                Text(message)
+                    .font(.caption2)
+                    .foregroundColor(.secondary)
+                    .fixedSize(horizontal: false, vertical: true)
+            }
+        }
+    }
+
+    // MARK: - Actions
+
+    private func sendBugReport() async {
+        bugReportInFlight = true
+        bugReportMessage = "Collecting logs..."
+        let service = BugReportService()
+        do {
+            let outcome = try await service.sendReport(isManual: true)
+            bugReportMessage = outcome.message
+        } catch {
+            bugReportMessage = error.localizedDescription
+        }
+        bugReportInFlight = false
+    }
+
+    private func showUninstallConfirmationAlert() {
+        let alert = NSAlert()
+        alert.messageText = "Uninstall EXO"
+        alert.informativeText = """
+            This will remove EXO and all its system components:
+
+            • Network configuration daemon
+            • Launch at login registration
+            • EXO network location
+
+            The app will be moved to Trash.
+            """
+        alert.alertStyle = .warning
+        alert.addButton(withTitle: "Uninstall")
+        alert.addButton(withTitle: "Cancel")
+
+        if let uninstallButton = alert.buttons.first {
+            uninstallButton.hasDestructiveAction = true
+        }
+
+        let response = alert.runModal()
+        if response == .alertFirstButtonReturn {
+            performUninstall()
+        }
+    }
+
+    private func performUninstall() {
+        uninstallInProgress = true
+
+        controller.cancelPendingLaunch()
+        controller.stop()
+        stateService.stopPolling()
+
+        DispatchQueue.global(qos: .utility).async {
+            do {
+                try NetworkSetupHelper.uninstall()
+
+                DispatchQueue.main.async {
+                    LaunchAtLoginHelper.disable()
+                    self.moveAppToTrash()
+
+                    DispatchQueue.main.asyncAfter(deadline: .now() + 0.5) {
+                        NSApplication.shared.terminate(nil)
+                    }
+                }
+            } catch {
+                DispatchQueue.main.async {
+                    let errorAlert = NSAlert()
+                    errorAlert.messageText = "Uninstall Failed"
+                    errorAlert.informativeText = error.localizedDescription
+                    errorAlert.alertStyle = .critical
+                    errorAlert.addButton(withTitle: "OK")
+                    errorAlert.runModal()
+                    self.uninstallInProgress = false
+                }
+            }
+        }
+    }
+
+    private func moveAppToTrash() {
+        guard let appURL = Bundle.main.bundleURL as URL? else { return }
+        do {
+            try FileManager.default.trashItem(at: appURL, resultingItemURL: nil)
+        } catch {
+            // If we can't trash the app, that's OK - user can do it manually
+        }
+    }
+
+    // MARK: - Helpers
+
+    private var hasGeneralChanges: Bool {
+        pendingNamespace != controller.customNamespace || pendingHFToken != controller.hfToken
+    }
+
+    private var hasModelChanges: Bool {
+        pendingEnableImageModels != controller.enableImageModels
+    }
+
+    private func applyGeneralSettings() {
+        controller.customNamespace = pendingNamespace
+        controller.hfToken = pendingHFToken
+        restartIfRunning()
+    }
+
+    private func applyModelSettings() {
+        controller.enableImageModels = pendingEnableImageModels
+        restartIfRunning()
+    }
+
+    private func restartIfRunning() {
+        if controller.status == .running || controller.status == .starting {
+            controller.restart()
+        }
+    }
+
+    private var buildTag: String {
+        Bundle.main.infoDictionary?["EXOBuildTag"] as? String ?? "unknown"
+    }
+
+    private var buildCommit: String {
+        Bundle.main.infoDictionary?["EXOBuildCommit"] as? String ?? "unknown"
+    }
+}

app/EXO/EXO/Views/SettingsWindowController.swift

@@ -0,0 +1,47 @@
+import AppKit
+import SwiftUI
+
+/// Manages a standalone native macOS Settings window.
+/// Ensures only one instance exists and brings it to front on repeated opens.
+@MainActor
+final class SettingsWindowController: ObservableObject {
+    private var window: NSWindow?
+
+    func open(
+        controller: ExoProcessController,
+        updater: SparkleUpdater,
+        networkStatusService: NetworkStatusService,
+        thunderboltBridgeService: ThunderboltBridgeService,
+        stateService: ClusterStateService
+    ) {
+        if let existing = window, existing.isVisible {
+            existing.makeKeyAndOrderFront(nil)
+            NSApp.activate(ignoringOtherApps: true)
+            return
+        }
+
+        let settingsView = SettingsView()
+            .environmentObject(controller)
+            .environmentObject(updater)
+            .environmentObject(networkStatusService)
+            .environmentObject(thunderboltBridgeService)
+            .environmentObject(stateService)
+
+        let hostingView = NSHostingView(rootView: settingsView)
+
+        let newWindow = NSWindow(
+            contentRect: NSRect(x: 0, y: 0, width: 450, height: 400),
+            styleMask: [.titled, .closable],
+            backing: .buffered,
+            defer: false
+        )
+        newWindow.title = "EXO Settings"
+        newWindow.contentView = hostingView
+        newWindow.center()
+        newWindow.isReleasedWhenClosed = false
+        newWindow.makeKeyAndOrderFront(nil)
+        NSApp.activate(ignoringOtherApps: true)
+
+        window = newWindow
+    }
+}

bench/exo_bench.py

@@ -75,7 +75,7 @@ def load_tokenizer_for_bench(model_id: str) -> Any:
         model_path = Path(
             snapshot_download(
                 model_id,
-                allow_patterns=["*.json", "*.py", "*.tiktoken"],
+                allow_patterns=["*.json", "*.py", "*.tiktoken", "*.model"],
             )
         )
 

bench/harness.py

@@ -371,7 +371,7 @@ def run_planning_phase(
                 unwrap_instance(p["DownloadCompleted"]["shardMetadata"])["modelCard"][
                     "modelId"
                 ],
-                p["DownloadCompleted"]["totalBytes"]["inBytes"],
+                p["DownloadCompleted"]["total"]["inBytes"],
             )
             for p in node_downloads
             if "DownloadCompleted" in p

bench/pyproject.toml

@@ -10,6 +10,7 @@ dependencies = [
     "huggingface-hub>=0.33.4",
     "tiktoken>=0.12.0",
     "jinja2>=3.1.0",
+    "protobuf>=5.29.0",
 ]
 
 [build-system]

dashboard/src/app.css

@@ -202,6 +202,23 @@
 	filter: drop-shadow(0 0 3px oklch(0.85 0.18 85 / 0.5));
 }
 
+/* Onboarding step 2: connection line between devices */
+.onboarding-connection-line {
+	stroke: oklch(0.85 0.18 85 / 0.5);
+	stroke-width: 1.5px;
+	stroke-dasharray: 6, 6;
+	animation: flowAnimation 1s linear infinite;
+	filter: drop-shadow(0 0 4px oklch(0.85 0.18 85 / 0.4));
+}
+
+/* Onboarding step 4: red connection line for disconnect */
+.onboarding-connection-line-red {
+	stroke: rgba(220, 38, 38, 0.7);
+	stroke-width: 1.5px;
+	stroke-dasharray: 6, 6;
+	filter: drop-shadow(0 0 2px rgba(220, 38, 38, 0.3));
+}
+
 .graph-link-active {
 	stroke: oklch(0.85 0.18 85 / 0.8);
 	stroke-width: 2px;
@@ -320,3 +337,51 @@ input:focus, textarea:focus {
 		transform: translate(400px, 400px);
 	}
 }
+
+/* Onboarding smooth fade-in */
+@keyframes onb-fade-in {
+	from { opacity: 0; transform: translateY(8px); }
+	to { opacity: 1; transform: none; }
+}
+
+/* Opacity-only fade-in — no transform, safe for containers with fixed-position children */
+@keyframes onb-fade-opacity {
+	from { opacity: 0; }
+	to { opacity: 1; }
+}
+
+/* Respect reduced motion preference */
+@media (prefers-reduced-motion: reduce) {
+	.shooting-star,
+	.shooting-star::before {
+		animation: none !important;
+		opacity: 0 !important;
+	}
+	.graph-link {
+		animation: none;
+	}
+	.status-pulse {
+		animation: none;
+	}
+	.cursor-blink {
+		animation: none;
+	}
+	.onboarding-connection-line {
+		animation: none;
+	}
+	.onboarding-connection-line-red {
+		animation: none;
+	}
+	[style*="onb-fade-in"],
+	[style*="onb-fade-opacity"] {
+		animation: none !important;
+		opacity: 1 !important;
+	}
+	*,
+	*::before,
+	*::after {
+		transition-duration: 0.01ms !important;
+		animation-duration: 0.01ms !important;
+		animation-iteration-count: 1 !important;
+	}
+}

dashboard/src/lib/components/ChatForm.svelte

@@ -7,8 +7,6 @@
     editingImage,
     clearEditingImage,
     selectedChatModel,
-    setSelectedChatModel,
-    instances,
     ttftMs,
     tps,
     totalTokens,
@@ -29,6 +27,19 @@
     showModelSelector?: boolean;
     modelTasks?: Record<string, string[]>;
     modelCapabilities?: Record<string, string[]>;
+    onSend?: () => void;
+    onAutoSend?: (
+      content: string,
+      files?: {
+        id: string;
+        name: string;
+        type: string;
+        textContent?: string;
+        preview?: string;
+      }[],
+    ) => void;
+    onOpenModelPicker?: () => void;
+    modelDisplayOverride?: string;
   }
 
   let {
@@ -39,6 +50,10 @@
     showModelSelector = false,
     modelTasks = {},
     modelCapabilities = {},
+    onSend,
+    onAutoSend,
+    onOpenModelPicker,
+    modelDisplayOverride,
   }: Props = $props();
 
   let message = $state("");
@@ -49,27 +64,12 @@
   const thinkingEnabled = $derived(thinkingEnabledStore());
   let loading = $derived(isLoading());
   const currentModel = $derived(selectedChatModel());
-  const instanceData = $derived(instances());
   const currentTtft = $derived(ttftMs());
   const currentTps = $derived(tps());
   const currentTokens = $derived(totalTokens());
   const currentEditingImage = $derived(editingImage());
   const isEditMode = $derived(currentEditingImage !== null);
 
-  // Custom dropdown state
-  let isModelDropdownOpen = $state(false);
-  let dropdownButtonRef: HTMLButtonElement | undefined = $state();
-  let dropdownPosition = $derived(() => {
-    if (!dropdownButtonRef || !isModelDropdownOpen)
-      return { top: 0, left: 0, width: 0 };
-    const rect = dropdownButtonRef.getBoundingClientRect();
-    return {
-      top: rect.top,
-      left: rect.left,
-      width: rect.width,
-    };
-  });
-
   // Accept all supported file types
   const acceptString = getAcceptString(["image", "text", "pdf"]);
 
@@ -122,73 +122,14 @@
         uploadedFiles.length > 0),
   );
 
-  // Extract available models from running instances
-  const availableModels = $derived(() => {
-    const models: Array<{ id: string; label: string; isImageModel: boolean }> =
-      [];
-    for (const [, instance] of Object.entries(instanceData)) {
-      const modelId = getInstanceModelId(instance);
-      if (
-        modelId &&
-        modelId !== "Unknown" &&
-        !models.some((m) => m.id === modelId)
-      ) {
-        models.push({
-          id: modelId,
-          label: modelId.split("/").pop() || modelId,
-          isImageModel: modelSupportsImageGeneration(modelId),
-        });
-      }
-    }
-    return models;
-  });
-
-  // Track previous model IDs to detect newly added models (plain variable to avoid reactive loop)
-  let previousModelIds: Set<string> = new Set();
-
-  // Auto-select the first available model if none is selected, if current selection is stale, or if a new model is added
-  $effect(() => {
-    const models = availableModels();
-    const currentModelIds = new Set(models.map((m) => m.id));
-
-    if (models.length > 0) {
-      // Find newly added models (in current but not in previous)
-      const newModels = models.filter((m) => !previousModelIds.has(m.id));
-
-      // If no model selected, select the first available
-      if (!currentModel) {
-        setSelectedChatModel(models[0].id);
-      }
-      // If current model is stale (no longer has a running instance), reset to first available
-      else if (!models.some((m) => m.id === currentModel)) {
-        setSelectedChatModel(models[0].id);
-      }
-      // If a new model was just added, select it
-      else if (newModels.length > 0 && previousModelIds.size > 0) {
-        setSelectedChatModel(newModels[0].id);
-      }
-    } else {
-      // No instances running - clear the selected model
-      if (currentModel) {
-        setSelectedChatModel("");
-      }
-    }
-
-    // Update previous model IDs for next comparison
-    previousModelIds = currentModelIds;
-  });
-
-  function getInstanceModelId(instanceWrapped: unknown): string {
-    if (!instanceWrapped || typeof instanceWrapped !== "object") return "";
-    const keys = Object.keys(instanceWrapped as Record<string, unknown>);
-    if (keys.length === 1) {
-      const instance = (instanceWrapped as Record<string, unknown>)[
-        keys[0]
-      ] as { shardAssignments?: { modelId?: string } };
-      return instance?.shardAssignments?.modelId || "";
-    }
-    return "";
-  }
+  // Short label for the currently selected model
+  const currentModelLabel = $derived(
+    currentModel
+      ? currentModel.split("/").pop() || currentModel
+      : modelDisplayOverride
+        ? modelDisplayOverride.split("/").pop() || modelDisplayOverride
+        : "",
+  );
 
   async function handleFiles(files: File[]) {
     if (files.length === 0) return;
@@ -275,6 +216,15 @@
     uploadedFiles = [];
     resetTextareaHeight();
 
+    // When onAutoSend is provided, the parent controls all send logic
+    // (including launching non-running models before sending)
+    if (onAutoSend) {
+      onAutoSend(content, files);
+      onSend?.();
+      setTimeout(() => textareaRef?.focus(), 10);
+      return;
+    }
+
     // Use image editing if in edit mode
     if (isEditMode && currentEditingImage && content) {
       editImage(content, currentEditingImage.imageDataUrl);
@@ -306,6 +256,8 @@
       );
     }
 
+    onSend?.();
+
     // Refocus the textarea after sending
     setTimeout(() => textareaRef?.focus(), 10);
   }
@@ -416,7 +368,7 @@
     {/if}
 
     <!-- Model selector (when enabled) -->
-    {#if showModelSelector && availableModels().length > 0}
+    {#if showModelSelector}
       <div
         class="flex items-center justify-between gap-2 px-3 py-2 border-b border-exo-medium-gray/30"
       >
@@ -425,33 +377,22 @@
             class="text-xs text-exo-light-gray uppercase tracking-wider flex-shrink-0"
             >MODEL:</span
           >
-          <!-- Custom dropdown -->
+          <!-- Model button — opens the full model picker -->
           <div class="relative flex-1 max-w-xs">
             <button
-              bind:this={dropdownButtonRef}
               type="button"
-              onclick={() => (isModelDropdownOpen = !isModelDropdownOpen)}
-              class="w-full bg-exo-medium-gray/50 border border-exo-yellow/30 rounded pl-3 pr-8 py-1.5 text-xs font-mono text-left tracking-wide cursor-pointer transition-all duration-200 hover:border-exo-yellow/50 focus:outline-none focus:border-exo-yellow/70 {isModelDropdownOpen
-                ? 'border-exo-yellow/70'
-                : ''}"
+              onclick={() => onOpenModelPicker?.()}
+              class="w-full bg-exo-medium-gray/50 border border-exo-yellow/30 rounded pl-3 pr-8 py-1.5 text-xs font-mono text-left tracking-wide cursor-pointer transition-all duration-200 hover:border-exo-yellow/50 focus:outline-none focus:border-exo-yellow/70"
             >
-              {#if availableModels().find((m) => m.id === currentModel)}
-                <span class="text-exo-yellow truncate"
-                  >{availableModels().find((m) => m.id === currentModel)
-                    ?.label}</span
-                >
-              {:else if availableModels().length > 0}
-                <span class="text-exo-yellow truncate"
-                  >{availableModels()[0].label}</span
+              {#if currentModelLabel}
+                <span class="text-exo-yellow truncate">{currentModelLabel}</span
                 >
               {:else}
                 <span class="text-exo-light-gray/50">— SELECT MODEL —</span>
               {/if}
             </button>
             <div
-              class="absolute right-2 top-1/2 -translate-y-1/2 pointer-events-none transition-transform duration-200 {isModelDropdownOpen
-                ? 'rotate-180'
-                : ''}"
+              class="absolute right-2 top-1/2 -translate-y-1/2 pointer-events-none"
             >
               <svg
                 class="w-3 h-3 text-exo-yellow/60"
@@ -468,78 +409,6 @@
               </svg>
             </div>
           </div>
-
-          {#if isModelDropdownOpen}
-            <!-- Backdrop to close dropdown -->
-            <button
-              type="button"
-              class="fixed inset-0 z-[9998] cursor-default"
-              onclick={() => (isModelDropdownOpen = false)}
-              aria-label="Close dropdown"
-            ></button>
-
-            <!-- Dropdown Panel - fixed positioning to escape overflow:hidden -->
-            <div
-              class="fixed bg-exo-dark-gray border border-exo-yellow/30 rounded shadow-lg shadow-black/50 z-[9999] max-h-48 overflow-y-auto"
-              style="bottom: calc(100vh - {dropdownPosition()
-                .top}px + 4px); left: {dropdownPosition()
-                .left}px; width: {dropdownPosition().width}px;"
-            >
-              <div class="py-1">
-                {#each availableModels() as model}
-                  <button
-                    type="button"
-                    onclick={() => {
-                      setSelectedChatModel(model.id);
-                      isModelDropdownOpen = false;
-                    }}
-                    class="w-full px-3 py-2 text-left text-xs font-mono tracking-wide transition-colors duration-100 flex items-center gap-2 {currentModel ===
-                    model.id
-                      ? 'bg-transparent text-exo-yellow'
-                      : 'text-exo-light-gray hover:text-exo-yellow'}"
-                  >
-                    {#if currentModel === model.id}
-                      <svg
-                        class="w-3 h-3 flex-shrink-0"
-                        fill="currentColor"
-                        viewBox="0 0 20 20"
-                      >
-                        <path
-                          fill-rule="evenodd"
-                          d="M16.707 5.293a1 1 0 010 1.414l-8 8a1 1 0 01-1.414 0l-4-4a1 1 0 011.414-1.414L8 12.586l7.293-7.293a1 1 0 011.414 0z"
-                          clip-rule="evenodd"
-                        />
-                      </svg>
-                    {:else}
-                      <span class="w-3"></span>
-                    {/if}
-                    {#if model.isImageModel}
-                      <svg
-                        class="w-3.5 h-3.5 flex-shrink-0 text-exo-yellow"
-                        fill="none"
-                        viewBox="0 0 24 24"
-                        stroke="currentColor"
-                        stroke-width="2"
-                        aria-label="Image generation model"
-                      >
-                        <rect
-                          x="3"
-                          y="3"
-                          width="18"
-                          height="18"
-                          rx="2"
-                          ry="2"
-                        />
-                        <circle cx="8.5" cy="8.5" r="1.5" />
-                        <polyline points="21 15 16 10 5 21" />
-                      </svg>
-                    {/if}
-                    <span class="truncate flex-1">{model.label}</span>
-                  </button>
-                {/each}
-              </div>
-            </div>
-          {/if}
         </div>
         <!-- Thinking toggle -->
         {#if modelSupportsThinking()}

dashboard/src/lib/components/ChatMessages.svelte

@@ -807,8 +807,8 @@
       >
         AWAITING INPUT
       </p>
-      <p class="text-sm sm:text-xs text-exo-light-gray tracking-wider mt-1">
-        ENTER A QUERY TO BEGIN
+      <p class="text-xs text-white/30 tracking-wider mt-1.5 font-mono">
+        Type a message below &middot; Shift+Enter for newline
       </p>
     </div>
   {/if}
@@ -823,6 +823,7 @@
       onclick={scrollToBottom}
       class="sticky bottom-4 left-1/2 -translate-x-1/2 w-10 h-10 rounded-full bg-exo-dark-gray/90 border border-exo-medium-gray/50 flex items-center justify-center text-exo-light-gray hover:text-exo-yellow hover:border-exo-yellow/50 transition-all shadow-lg cursor-pointer z-10"
       title="Scroll to bottom"
+      aria-label="Scroll to bottom of messages"
     >
       <svg
         class="w-5 h-5"

dashboard/src/lib/components/ChatModelSelector.svelte

@@ -0,0 +1,401 @@
+<script lang="ts" module>
+  export interface ChatModelInfo {
+    id: string;
+    name: string;
+    base_model: string;
+    storage_size_megabytes: number;
+    capabilities: string[];
+    family: string;
+    quantization: string;
+  }
+
+  // Auto mode tier list (for when user just starts typing)
+  export const AUTO_TIERS: string[][] = [
+    // Tier 1 (frontier)
+    ["DeepSeek V3.1", "GLM-5", "Kimi K2.5", "Qwen3 Coder Next"],
+    // Tier 2 (excellent)
+    [
+      "Kimi K2",
+      "Qwen3 235B",
+      "MiniMax M2.5",
+      "Step 3.5 Flash",
+      "Qwen3 Next 80B",
+    ],
+    // Tier 3 (great)
+    [
+      "GLM 4.7",
+      "MiniMax M2.1",
+      "Qwen3 Coder 480B",
+      "GLM 4.5 Air",
+      "Llama 3.3 70B",
+    ],
+    // Tier 4 (good)
+    ["GPT-OSS 120B", "Qwen3 30B", "Llama 3.1 70B", "GLM 4.7 Flash"],
+    // Tier 5 (small/fast)
+    [
+      "Llama 3.1 8B",
+      "GPT-OSS 20B",
+      "Llama 3.2 3B",
+      "Qwen3 0.6B",
+      "Llama 3.2 1B",
+    ],
+  ];
+
+  /** Return the tier index (0 = best) for a base_model name. */
+  export function getAutoTierIndex(baseModel: string): number {
+    for (let i = 0; i < AUTO_TIERS.length; i++) {
+      if (AUTO_TIERS[i].includes(baseModel)) return i;
+    }
+    return AUTO_TIERS.length; // not in any tier → lowest priority
+  }
+
+  /** Auto mode: walk tiers top-down, pick biggest fitting variant from highest tier. */
+  export function pickAutoModel(
+    modelList: ChatModelInfo[],
+    memoryGB: number,
+  ): ChatModelInfo | null {
+    for (const tier of AUTO_TIERS) {
+      const candidates: ChatModelInfo[] = [];
+      for (const baseModel of tier) {
+        const variants = modelList
+          .filter(
+            (m) =>
+              m.base_model === baseModel &&
+              (m.storage_size_megabytes || 0) / 1024 <= memoryGB &&
+              (m.storage_size_megabytes || 0) > 0,
+          )
+          .sort(
+            (a, b) =>
+              (b.storage_size_megabytes || 0) - (a.storage_size_megabytes || 0),
+          );
+        if (variants[0]) candidates.push(variants[0]);
+      }
+      if (candidates.length > 0) {
+        candidates.sort(
+          (a, b) =>
+            (b.storage_size_megabytes || 0) - (a.storage_size_megabytes || 0),
+        );
+        return candidates[0];
+      }
+    }
+    return null;
+  }
+</script>
+
+<script lang="ts">
+  interface CategoryRecommendation {
+    category: "coding" | "writing" | "agentic" | "biggest";
+    label: string;
+    model: ChatModelInfo | null;
+    tooltip: string;
+  }
+
+  interface Props {
+    models: ChatModelInfo[];
+    clusterLabel: string;
+    totalMemoryGB: number;
+    onSelect: (modelId: string, category: string) => void;
+    onAddModel: () => void;
+    class?: string;
+  }
+
+  let {
+    models,
+    clusterLabel,
+    totalMemoryGB,
+    onSelect,
+    onAddModel,
+    class: className = "",
+  }: Props = $props();
+
+  // --- Hardcoded Rankings ---
+  const CODING_RANKING = [
+    "Qwen3 Coder Next",
+    "Qwen3 Coder 480B",
+    "Qwen3 30B",
+    "GPT-OSS 20B",
+    "Llama 3.1 8B",
+    "Llama 3.2 3B",
+    "Qwen3 0.6B",
+  ];
+
+  const WRITING_RANKING = [
+    "Kimi K2.5",
+    "Kimi K2",
+    "Qwen3 Next 80B",
+    "Llama 3.3 70B",
+    "MiniMax M2.5",
+    "GLM 4.5 Air",
+    "GLM 4.7 Flash",
+    "GPT-OSS 20B",
+    "Llama 3.1 8B",
+    "Llama 3.2 3B",
+    "Qwen3 0.6B",
+  ];
+
+  const AGENTIC_RANKING = [
+    "DeepSeek V3.1",
+    "GLM-5",
+    "Qwen3 235B",
+    "Step 3.5 Flash",
+    "GLM 4.7",
+    "MiniMax M2.1",
+    "GPT-OSS 120B",
+    "Llama 3.3 70B",
+    "Llama 3.1 70B",
+    "GLM 4.7 Flash",
+    "GPT-OSS 20B",
+    "Qwen3 30B",
+    "Llama 3.1 8B",
+    "Llama 3.2 3B",
+    "Qwen3 0.6B",
+  ];
+
+  function getModelSizeGB(m: ChatModelInfo): number {
+    return (m.storage_size_megabytes || 0) / 1024;
+  }
+
+  function fitsInMemory(m: ChatModelInfo): boolean {
+    return getModelSizeGB(m) <= totalMemoryGB && getModelSizeGB(m) > 0;
+  }
+
+  /** For a given base_model name, find the biggest quant variant that fits in memory. */
+  function pickBestVariant(baseModel: string): ChatModelInfo | null {
+    const variants = models
+      .filter((m) => m.base_model === baseModel && fitsInMemory(m))
+      .sort((a, b) => getModelSizeGB(b) - getModelSizeGB(a));
+    return variants[0] ?? null;
+  }
+
+  /** Walk a ranked list of base_model names, return the first that has a fitting variant. */
+  function pickFromRanking(ranking: string[]): ChatModelInfo | null {
+    for (const baseModel of ranking) {
+      const pick = pickBestVariant(baseModel);
+      if (pick) return pick;
+    }
+    return null;
+  }
+
+  /** Pick the single biggest model that fits. */
+  function pickBiggest(): ChatModelInfo | null {
+    const fitting = models
+      .filter((m) => fitsInMemory(m))
+      .sort((a, b) => getModelSizeGB(b) - getModelSizeGB(a));
+    return fitting[0] ?? null;
+  }
+
+  const recommendations = $derived.by((): CategoryRecommendation[] => {
+    return [
+      {
+        category: "coding",
+        label: "Best for Coding",
+        model: pickFromRanking(CODING_RANKING),
+        tooltip:
+          "Ranked by coding benchmark performance (LiveCodeBench, SWE-bench)",
+      },
+      {
+        category: "writing",
+        label: "Best for Writing",
+        model: pickFromRanking(WRITING_RANKING),
+        tooltip: "Ranked by creative writing quality and instruction following",
+      },
+      {
+        category: "agentic",
+        label: "Best Agentic",
+        model: pickFromRanking(AGENTIC_RANKING),
+        tooltip: "Ranked by reasoning, planning, and tool-use capability",
+      },
+      {
+        category: "biggest",
+        label: "Biggest",
+        model: pickBiggest(),
+        tooltip: "Largest model that fits in your available memory",
+      },
+    ];
+  });
+
+  function formatSize(mb: number): string {
+    const gb = mb / 1024;
+    if (gb >= 100) return `${Math.round(gb)} GB`;
+    return `${gb.toFixed(1)} GB`;
+  }
+
+  // Category icons (SVG paths)
+  const categoryIcons: Record<string, string> = {
+    coding:
+      "M8 9l3 3-3 3m5 0h3M5 20h14a2 2 0 002-2V6a2 2 0 00-2-2H5a2 2 0 00-2 2v12a2 2 0 002 2z",
+    writing:
+      "M11 5H6a2 2 0 00-2 2v11a2 2 0 002 2h11a2 2 0 002-2v-5m-1.414-9.414a2 2 0 112.828 2.828L11.828 15H9v-2.828l8.586-8.586z",
+    agentic:
+      "M9.663 17h4.673M12 3v1m6.364 1.636l-.707.707M21 12h-1M4 12H3m3.343-5.657l-.707-.707m2.828 9.9a5 5 0 117.072 0l-.548.547A3.374 3.374 0 0014 18.469V19a2 2 0 11-4 0v-.531c0-.895-.356-1.754-.988-2.386l-.548-.547z",
+    biggest:
+      "M19 11H5m14 0a2 2 0 012 2v6a2 2 0 01-2 2H5a2 2 0 01-2-2v-6a2 2 0 012-2m14 0V9a2 2 0 00-2-2M5 11V9a2 2 0 012-2m0 0V5a2 2 0 012-2h6a2 2 0 012 2v2M7 7h10",
+  };
+
+  let hoveredTooltip = $state<string | null>(null);
+  let tooltipAnchor = $state<{ x: number; y: number } | null>(null);
+
+  function showTooltip(category: string, e: MouseEvent | FocusEvent) {
+    hoveredTooltip = category;
+    const target = e.currentTarget as HTMLElement;
+    const rect = target.getBoundingClientRect();
+    tooltipAnchor = { x: rect.left + rect.width / 2, y: rect.top };
+  }
+
+  function hideTooltip() {
+    hoveredTooltip = null;
+    tooltipAnchor = null;
+  }
+</script>
+
+<div class="flex flex-col items-center justify-center gap-6 {className}">
+  <!-- Header -->
+  <div class="text-center">
+    <p class="text-xs text-exo-light-gray uppercase tracking-[0.2em] mb-1">
+      Recommended for your
+    </p>
+    <p class="text-sm text-white font-mono tracking-wide">{clusterLabel}</p>
+  </div>
+
+  <!-- Category Cards Grid -->
+  <div class="grid grid-cols-2 gap-3 w-full max-w-md">
+    {#each recommendations as rec}
+      {#if rec.model}
+        <button
+          type="button"
+          onclick={() => rec.model && onSelect(rec.model.id, rec.category)}
+          class="group relative flex flex-col items-start gap-2 p-4 rounded-lg border border-exo-medium-gray/50 bg-exo-dark-gray/50 hover:border-exo-yellow/40 hover:bg-exo-dark-gray transition-all duration-200 cursor-pointer text-left"
+        >
+          <!-- Category icon + label -->
+          <div class="flex items-center gap-2 w-full">
+            <svg
+              class="w-4 h-4 text-exo-yellow/70 group-hover:text-exo-yellow transition-colors flex-shrink-0"
+              fill="none"
+              viewBox="0 0 24 24"
+              stroke="currentColor"
+              stroke-width="1.5"
+            >
+              <path
+                stroke-linecap="round"
+                stroke-linejoin="round"
+                d={categoryIcons[rec.category]}
+              />
+            </svg>
+            <span
+              class="text-xs font-mono uppercase tracking-wider text-exo-light-gray group-hover:text-white transition-colors"
+            >
+              {rec.label}
+            </span>
+            <!-- Info tooltip -->
+            <div class="ml-auto flex-shrink-0">
+              <span
+                role="button"
+                tabindex="-1"
+                class="text-exo-light-gray/40 hover:text-exo-light-gray transition-colors cursor-help inline-flex"
+                onmouseenter={(e: MouseEvent) => showTooltip(rec.category, e)}
+                onmouseleave={() => hideTooltip()}
+                onclick={(e: MouseEvent) => {
+                  e.stopPropagation();
+                  if (hoveredTooltip === rec.category) {
+                    hideTooltip();
+                  } else {
+                    showTooltip(rec.category, e);
+                  }
+                }}
+              >
+                <svg
+                  class="w-3.5 h-3.5"
+                  fill="none"
+                  viewBox="0 0 24 24"
+                  stroke="currentColor"
+                  stroke-width="2"
+                >
+                  <circle cx="12" cy="12" r="10" />
+                  <path d="M12 16v-4m0-4h.01" />
+                </svg>
+              </span>
+            </div>
+          </div>
+
+          <!-- Model name + size -->
+          <div class="w-full">
+            <p class="text-sm text-white font-mono truncate">
+              {rec.model.base_model}
+            </p>
+            <p class="text-xs text-exo-light-gray/60 font-mono mt-0.5">
+              {formatSize(rec.model.storage_size_megabytes)}
+              {#if rec.model.quantization}
+                <span class="text-exo-light-gray/40"
+                  >&middot; {rec.model.quantization}</span
+                >
+              {/if}
+            </p>
+          </div>
+        </button>
+      {:else}
+        <!-- No model fits for this category -->
+        <div
+          class="flex flex-col items-start gap-2 p-4 rounded-lg border border-exo-medium-gray/30 bg-exo-dark-gray/30 opacity-50"
+        >
+          <div class="flex items-center gap-2">
+            <svg
+              class="w-4 h-4 text-exo-light-gray/40 flex-shrink-0"
+              fill="none"
+              viewBox="0 0 24 24"
+              stroke="currentColor"
+              stroke-width="1.5"
+            >
+              <path
+                stroke-linecap="round"
+                stroke-linejoin="round"
+                d={categoryIcons[rec.category]}
+              />
+            </svg>
+            <span
+              class="text-xs font-mono uppercase tracking-wider text-exo-light-gray/50"
+              >{rec.label}</span
+            >
+          </div>
+          <p class="text-xs text-exo-light-gray/40 font-mono">No model fits</p>
+        </div>
+      {/if}
+    {/each}
+  </div>
+
+  <!-- Add Model Button -->
+  <button
+    type="button"
+    onclick={onAddModel}
+    class="flex items-center gap-2 px-4 py-2 text-xs font-mono uppercase tracking-wider text-exo-light-gray hover:text-exo-yellow border border-exo-medium-gray/30 hover:border-exo-yellow/30 rounded-lg transition-all duration-200 cursor-pointer"
+  >
+    <svg
+      class="w-3.5 h-3.5"
+      fill="none"
+      viewBox="0 0 24 24"
+      stroke="currentColor"
+      stroke-width="2"
+    >
+      <path stroke-linecap="round" stroke-linejoin="round" d="M12 4v16m8-8H4" />
+    </svg>
+    Add Model
+  </button>
+
+  <!-- Auto hint -->
+  <p class="text-xs text-exo-light-gray/40 font-mono tracking-wide text-center">
+    Or just start typing &mdash; we'll pick the best model automatically
+  </p>
+</div>
+
+<!-- Fixed-position tooltip (escapes overflow-hidden ancestors) -->
+{#if hoveredTooltip && tooltipAnchor}
+  {@const rec = recommendations.find((r) => r.category === hoveredTooltip)}
+  {#if rec}
+    <div
+      class="fixed z-[9999] px-3 py-2 bg-exo-black border border-exo-medium-gray/50 rounded text-xs text-exo-light-gray whitespace-nowrap shadow-lg pointer-events-none"
+      style="left: {tooltipAnchor.x}px; top: {tooltipAnchor.y -
+        8}px; transform: translate(-50%, -100%);"
+    >
+      {rec.tooltip}
+    </div>
+  {/if}
+{/if}

dashboard/src/lib/components/ChatSidebar.svelte

@@ -2,7 +2,6 @@
   import {
     conversations,
     activeConversationId,
-    createConversation,
     loadConversation,
     deleteConversation,
     deleteAllConversations,
@@ -17,9 +16,15 @@
 
   interface Props {
     class?: string;
+    onNewChat?: () => void;
+    onSelectConversation?: () => void;
   }
 
-  let { class: className = "" }: Props = $props();
+  let {
+    class: className = "",
+    onNewChat,
+    onSelectConversation,
+  }: Props = $props();
 
   const conversationList = $derived(conversations());
   const activeId = $derived(activeConversationId());
@@ -42,10 +47,11 @@
   );
 
   function handleNewChat() {
-    createConversation();
+    onNewChat?.();
   }
 
   function handleSelectConversation(id: string) {
+    onSelectConversation?.();
     loadConversation(id);
   }
 
@@ -185,11 +191,7 @@
 
     let instanceType: string | null = null;
     if (instanceTag === "MlxRingInstance") instanceType = "MLX Ring";
-    else if (
-      instanceTag === "MlxIbvInstance" ||
-      instanceTag === "MlxJacclInstance"
-    )
-      instanceType = "MLX RDMA";
+    else if (instanceTag === "MlxJacclInstance") instanceType = "MLX RDMA";
 
     let sharding: string | null = null;
     const inst = instance as {
@@ -307,7 +309,7 @@
       <div class="py-2">
         <div class="px-4 py-2">
           <span
-            class="text-sm text-white/70 font-mono tracking-wider uppercase"
+            class="text-xs text-exo-light-gray font-mono tracking-wider uppercase"
           >
             {searchQuery ? "SEARCH RESULTS" : "CONVERSATIONS"}
           </span>
@@ -376,39 +378,37 @@
                 onkeydown={(e) =>
                   e.key === "Enter" &&
                   handleSelectConversation(conversation.id)}
-                class="group w-full flex items-center justify-between p-2 rounded mb-1 transition-all text-left cursor-pointer
+                class="group w-full flex items-center justify-between p-2.5 rounded-lg mb-1 transition-all text-left cursor-pointer
 									{activeId === conversation.id
-                  ? 'bg-transparent border border-exo-yellow/30'
-                  : 'hover:border-exo-yellow/20 border border-transparent'}"
+                  ? 'bg-exo-yellow/5 border border-exo-yellow/30'
+                  : 'hover:bg-white/[0.03] hover:border-white/10 border border-transparent'}"
               >
                 <div class="flex-1 min-w-0 pr-2">
                   <div
-                    class="text-sm truncate {activeId === conversation.id
+                    class="text-sm font-medium truncate {activeId ===
+                    conversation.id
                       ? 'text-exo-yellow'
-                      : 'text-white/90'}"
+                      : 'text-white'}"
                   >
                     {conversation.name}
                   </div>
-                  <div class="text-sm text-white/50 mt-0.5">
+                  <div class="text-xs text-white/60 mt-0.5">
                     {formatDate(conversation.updatedAt)}
                   </div>
-                  <div class="text-sm text-white/70 truncate">
+                  <div class="text-xs text-exo-light-gray truncate">
                     {info.modelLabel}
                   </div>
-                  <div class="text-xs text-white/60 font-mono">
-                    Strategy: <span class="text-white/80"
-                      >{info.strategyLabel}</span
-                    >
-                  </div>
                   {#if stats}
-                    <div class="text-xs text-white/60 font-mono mt-1">
-                      {#if stats.ttftMs}<span class="text-white/40">TTFT</span>
-                        {stats.ttftMs.toFixed(
-                          0,
-                        )}ms{/if}{#if stats.ttftMs && stats.tps}<span
-                          class="text-white/30 mx-1.5">•</span
-                        >{/if}{#if stats.tps}{stats.tps.toFixed(1)}
-                        <span class="text-white/40">tok/s</span>{/if}
+                    <div class="text-xs text-white/70 font-mono mt-1">
+                      {#if stats.ttftMs}<span class="text-white/50">TTFT</span>
+                        <span class="text-exo-yellow/80"
+                          >{stats.ttftMs.toFixed(0)}ms</span
+                        >{/if}{#if stats.ttftMs && stats.tps}<span
+                          class="text-white/30 mx-1.5">·</span
+                        >{/if}{#if stats.tps}<span class="text-exo-yellow/80"
+                          >{stats.tps.toFixed(1)}</span
+                        >
+                        <span class="text-white/50">tok/s</span>{/if}
                     </div>
                   {/if}
                 </div>

dashboard/src/lib/components/ConnectionBanner.svelte

@@ -0,0 +1,20 @@
+<script lang="ts">
+  import { isConnected } from "$lib/stores/app.svelte";
+  import { slide } from "svelte/transition";
+
+  const connected = $derived(isConnected());
+</script>
+
+{#if !connected}
+  <div
+    transition:slide={{ duration: 200 }}
+    class="relative z-50 flex items-center justify-center gap-2 px-4 py-2 bg-red-950/80 border-b border-red-500/30"
+    role="alert"
+    aria-live="assertive"
+  >
+    <div class="w-2 h-2 bg-red-500 rounded-full animate-pulse"></div>
+    <span class="text-xs font-mono text-red-300 tracking-wider uppercase">
+      Connection lost &mdash; Reconnecting to backend&hellip;
+    </span>
+  </div>
+{/if}

dashboard/src/lib/components/DeviceIcon.svelte

@@ -0,0 +1,277 @@
+<script lang="ts">
+  /**
+   * DeviceIcon — renders a device icon as an SVG <g> element.
+   * Uses the exact same proportional math as TopologyGraph.svelte
+   * so that devices look identical in both the topology view and
+   * the onboarding animation.
+   *
+   * Must be placed inside an <svg> element.
+   */
+
+  interface Props {
+    /** "macbook pro" | "mac studio" | "mac mini" etc. */
+    deviceType: string;
+    /** Center X coordinate in SVG space */
+    cx: number;
+    /** Center Y coordinate in SVG space */
+    cy: number;
+    /** Base sizing factor (equivalent to TopologyGraph's nodeRadius) */
+    size?: number;
+    /** RAM usage 0–100 */
+    ramPercent?: number;
+    /** Unique id suffix for clip-path ids */
+    uid?: string;
+  }
+
+  let {
+    deviceType,
+    cx,
+    cy,
+    size = 60,
+    ramPercent = 60,
+    uid = "dev",
+  }: Props = $props();
+
+  // Apple logo path — same constant used by TopologyGraph
+  const APPLE_LOGO_PATH =
+    "M788.1 340.9c-5.8 4.5-108.2 62.2-108.2 190.5 0 148.4 130.3 200.9 134.2 202.2-.6 3.2-20.7 71.9-68.7 141.9-42.8 61.6-87.5 123.1-155.5 123.1s-85.5-39.5-164-39.5c-76.5 0-103.7 40.8-165.9 40.8s-105.6-57-155.5-127C46.7 790.7 0 663 0 541.8c0-194.4 126.4-297.5 250.8-297.5 66.1 0 121.2 43.4 162.7 43.4 39.5 0 101.1-46 176.3-46 28.5 0 130.9 2.6 198.3 99.2zm-234-181.5c31.1-36.9 53.1-88.1 53.1-139.3 0-7.1-.6-14.3-1.9-20.1-50.6 1.9-110.8 33.7-147.1 75.8-28.5 32.4-55.1 83.6-55.1 135.5 0 7.8 1.3 15.6 1.9 18.1 3.2.6 8.4 1.3 13.6 1.3 45.4 0 102.5-30.4 135.5-71.3z";
+  const LOGO_NATIVE_WIDTH = 814;
+  const LOGO_NATIVE_HEIGHT = 1000;
+
+  const wireColor = "rgba(179,179,179,0.8)";
+  const strokeWidth = 1.5;
+
+  const modelLower = $derived(deviceType.toLowerCase());
+
+  // ── Mac Studio dimensions (same ratios as TopologyGraph) ──
+  const studioW = $derived(size * 1.25);
+  const studioH = $derived(size * 0.85);
+  const studioX = $derived(cx - studioW / 2);
+  const studioY = $derived(cy - studioH / 2);
+  const studioCorner = 4;
+  const studioTopH = $derived(studioH * 0.15);
+
+  // Studio front panel details
+  const studioSlotH = $derived(studioH * 0.14);
+  const studioVSlotW = $derived(studioW * 0.05);
+  const studioVSlotY = $derived(
+    studioY + studioTopH + (studioH - studioTopH) * 0.6,
+  );
+  const studioVSlot1X = $derived(studioX + studioW * 0.18);
+  const studioVSlot2X = $derived(studioX + studioW * 0.28);
+  const studioHSlotW = $derived(studioW * 0.2);
+  const studioHSlotX = $derived(studioX + studioW * 0.5 - studioHSlotW / 2);
+
+  // Studio memory fill
+  const studioMemTotalH = $derived(studioH - studioTopH);
+  const studioMemH = $derived((ramPercent / 100) * studioMemTotalH);
+
+  // ── MacBook dimensions (same ratios as TopologyGraph) ──
+  const mbW = $derived((size * 1.6 * 0.85) / 1.15);
+  const mbH = $derived(size * 0.85);
+  const mbX = $derived(cx - mbW / 2);
+  const mbY = $derived(cy - mbH / 2);
+
+  const mbScreenH = $derived(mbH * 0.7);
+  const mbBaseH = $derived(mbH * 0.3);
+  const mbScreenW = $derived(mbW * 0.85);
+  const mbScreenX = $derived(cx - mbScreenW / 2);
+  const mbBezel = 3;
+
+  // MacBook memory fill
+  const mbMemTotalH = $derived(mbScreenH - mbBezel * 2);
+  const mbMemH = $derived((ramPercent / 100) * mbMemTotalH);
+
+  // Apple logo sizing
+  const mbLogoTargetH = $derived(mbScreenH * 0.22);
+  const mbLogoScale = $derived(mbLogoTargetH / LOGO_NATIVE_HEIGHT);
+  const mbLogoX = $derived(cx - (LOGO_NATIVE_WIDTH * mbLogoScale) / 2);
+  const mbLogoY = $derived(
+    mbY + mbScreenH / 2 - (LOGO_NATIVE_HEIGHT * mbLogoScale) / 2,
+  );
+
+  // MacBook base (trapezoidal)
+  const mbBaseY = $derived(mbY + mbScreenH);
+  const mbBaseTopW = $derived(mbScreenW);
+  const mbBaseBottomW = $derived(mbW);
+  const mbBaseTopX = $derived(cx - mbBaseTopW / 2);
+  const mbBaseBottomX = $derived(cx - mbBaseBottomW / 2);
+
+  // Keyboard
+  const mbKbX = $derived(mbBaseTopX + 6);
+  const mbKbY = $derived(mbBaseY + 3);
+  const mbKbW = $derived(mbBaseTopW - 12);
+  const mbKbH = $derived(mbBaseH * 0.55);
+
+  // Trackpad
+  const mbTpW = $derived(mbBaseTopW * 0.4);
+  const mbTpX = $derived(cx - mbTpW / 2);
+  const mbTpY = $derived(mbBaseY + mbKbH + 5);
+  const mbTpH = $derived(mbBaseH * 0.3);
+
+  // Clip IDs
+  const screenClipId = $derived(`di-screen-${uid}`);
+  const studioClipId = $derived(`di-studio-${uid}`);
+</script>
+
+{#if modelLower === "mac studio" || modelLower === "mac mini"}
+  <!-- Mac Studio / Mac Mini -->
+  <defs>
+    <clipPath id={studioClipId}>
+      <rect
+        x={studioX}
+        y={studioY + studioTopH}
+        width={studioW}
+        height={studioH - studioTopH}
+        rx={studioCorner - 1}
+      />
+    </clipPath>
+  </defs>
+
+  <!-- Main body -->
+  <rect
+    x={studioX}
+    y={studioY}
+    width={studioW}
+    height={studioH}
+    rx={studioCorner}
+    fill="#1a1a1a"
+    stroke={wireColor}
+    stroke-width={strokeWidth}
+  />
+
+  <!-- Memory fill -->
+  {#if ramPercent > 0}
+    <rect
+      x={studioX}
+      y={studioY + studioTopH + (studioMemTotalH - studioMemH)}
+      width={studioW}
+      height={studioMemH}
+      fill="rgba(255,215,0,0.75)"
+      clip-path="url(#{studioClipId})"
+    />
+  {/if}
+
+  <!-- Top surface divider -->
+  <line
+    x1={studioX}
+    y1={studioY + studioTopH}
+    x2={studioX + studioW}
+    y2={studioY + studioTopH}
+    stroke="rgba(179,179,179,0.3)"
+    stroke-width="0.5"
+  />
+
+  <!-- Front panel: vertical slots -->
+  <rect
+    x={studioVSlot1X - studioVSlotW / 2}
+    y={studioVSlotY}
+    width={studioVSlotW}
+    height={studioSlotH}
+    fill="rgba(0,0,0,0.35)"
+    rx="1.5"
+  />
+  <rect
+    x={studioVSlot2X - studioVSlotW / 2}
+    y={studioVSlotY}
+    width={studioVSlotW}
+    height={studioSlotH}
+    fill="rgba(0,0,0,0.35)"
+    rx="1.5"
+  />
+
+  <!-- Horizontal slot (SD card) -->
+  <rect
+    x={studioHSlotX}
+    y={studioVSlotY}
+    width={studioHSlotW}
+    height={studioSlotH * 0.6}
+    fill="rgba(0,0,0,0.35)"
+    rx="1"
+  />
+{:else}
+  <!-- MacBook Pro -->
+  <defs>
+    <clipPath id={screenClipId}>
+      <rect
+        x={mbScreenX + mbBezel}
+        y={mbY + mbBezel}
+        width={mbScreenW - mbBezel * 2}
+        height={mbScreenH - mbBezel * 2}
+        rx="2"
+      />
+    </clipPath>
+  </defs>
+
+  <!-- Screen outer frame -->
+  <rect
+    x={mbScreenX}
+    y={mbY}
+    width={mbScreenW}
+    height={mbScreenH}
+    rx="3"
+    fill="#1a1a1a"
+    stroke={wireColor}
+    stroke-width={strokeWidth}
+  />
+
+  <!-- Screen inner (dark) -->
+  <rect
+    x={mbScreenX + mbBezel}
+    y={mbY + mbBezel}
+    width={mbScreenW - mbBezel * 2}
+    height={mbScreenH - mbBezel * 2}
+    rx="2"
+    fill="#0a0a12"
+  />
+
+  <!-- Memory fill on screen -->
+  {#if ramPercent > 0}
+    <rect
+      x={mbScreenX + mbBezel}
+      y={mbY + mbBezel + (mbMemTotalH - mbMemH)}
+      width={mbScreenW - mbBezel * 2}
+      height={mbMemH}
+      fill="rgba(255,215,0,0.85)"
+      clip-path="url(#{screenClipId})"
+    />
+  {/if}
+
+  <!-- Apple logo -->
+  <path
+    d={APPLE_LOGO_PATH}
+    transform="translate({mbLogoX}, {mbLogoY}) scale({mbLogoScale})"
+    fill="#FFFFFF"
+    opacity="0.9"
+  />
+
+  <!-- Keyboard base (trapezoidal) -->
+  <path
+    d="M {mbBaseTopX} {mbBaseY} L {mbBaseTopX +
+      mbBaseTopW} {mbBaseY} L {mbBaseBottomX + mbBaseBottomW} {mbBaseY +
+      mbBaseH} L {mbBaseBottomX} {mbBaseY + mbBaseH} Z"
+    fill="#2c2c2c"
+    stroke={wireColor}
+    stroke-width="1"
+  />
+
+  <!-- Keyboard area -->
+  <rect
+    x={mbKbX}
+    y={mbKbY}
+    width={mbKbW}
+    height={mbKbH}
+    fill="rgba(0,0,0,0.2)"
+    rx="2"
+  />
+
+  <!-- Trackpad -->
+  <rect
+    x={mbTpX}
+    y={mbTpY}
+    width={mbTpW}
+    height={mbTpH}
+    fill="rgba(255,255,255,0.08)"
+    rx="2"
+  />
+{/if}

dashboard/src/lib/components/HeaderNav.svelte

@@ -6,6 +6,10 @@
   export let showSidebarToggle = false;
   export let sidebarVisible = true;
   export let onToggleSidebar: (() => void) | null = null;
+  export let downloadProgress: {
+    count: number;
+    percentage: number;
+  } | null = null;
 
   function handleHome(): void {
     if (onHome) {
@@ -33,13 +37,17 @@
     <div class="absolute left-6 top-1/2 -translate-y-1/2">
       <button
         onclick={handleToggleSidebar}
-        class="p-2 rounded border border-exo-medium-gray/40 hover:border-exo-yellow/50 transition-colors cursor-pointer"
+        class="p-2 rounded border border-exo-light-gray/30 hover:border-exo-yellow/50 hover:bg-exo-medium-gray/30 transition-colors cursor-pointer"
         title={sidebarVisible ? "Hide sidebar" : "Show sidebar"}
+        aria-label={sidebarVisible
+          ? "Hide conversation sidebar"
+          : "Show conversation sidebar"}
+        aria-pressed={sidebarVisible}
       >
         <svg
           class="w-5 h-5 {sidebarVisible
             ? 'text-exo-yellow'
-            : 'text-exo-medium-gray'}"
+            : 'text-exo-light-gray'}"
           fill="none"
           viewBox="0 0 24 24"
           stroke="currentColor"
@@ -75,13 +83,14 @@
     <img
       src="/exo-logo.png"
       alt="EXO"
-      class="h-18 drop-shadow-[0_0_20px_rgba(255,215,0,0.5)]"
+      class="h-18 drop-shadow-[0_0_4px_rgba(255,215,0,0.3)]"
     />
   </button>
 
   <!-- Right: Home + Downloads -->
-  <div
+  <nav
     class="absolute right-6 top-1/2 -translate-y-1/2 flex items-center gap-4"
+    aria-label="Main navigation"
   >
     {#if showHome}
       <button
@@ -110,20 +119,56 @@
       class="text-sm text-exo-light-gray hover:text-exo-yellow transition-colors tracking-wider uppercase flex items-center gap-2 cursor-pointer"
       title="View downloads overview"
     >
-      <svg
-        class="w-4 h-4"
-        viewBox="0 0 24 24"
-        fill="none"
-        stroke="currentColor"
-        stroke-width="2"
-        stroke-linecap="round"
-        stroke-linejoin="round"
-      >
-        <path d="M12 3v12" />
-        <path d="M7 12l5 5 5-5" />
-        <path d="M5 21h14" />
-      </svg>
+      {#if downloadProgress}
+        <!-- Compact download progress indicator -->
+        <div class="relative w-4 h-4 flex-shrink-0">
+          <svg class="w-4 h-4 -rotate-90" viewBox="0 0 20 20">
+            <circle
+              cx="10"
+              cy="10"
+              r="8"
+              fill="none"
+              stroke="currentColor"
+              stroke-width="2"
+              opacity="0.2"
+            />
+            <circle
+              cx="10"
+              cy="10"
+              r="8"
+              fill="none"
+              stroke="currentColor"
+              stroke-width="2"
+              stroke-dasharray={2 * Math.PI * 8}
+              stroke-dashoffset={2 *
+                Math.PI *
+                8 *
+                (1 - downloadProgress.percentage / 100)}
+              class="text-blue-400 transition-all duration-300"
+            />
+          </svg>
+          <div
+            class="absolute inset-0 flex items-center justify-center text-[6px] font-mono text-blue-400"
+          >
+            {downloadProgress.count}
+          </div>
+        </div>
+      {:else}
+        <svg
+          class="w-4 h-4"
+          viewBox="0 0 24 24"
+          fill="none"
+          stroke="currentColor"
+          stroke-width="2"
+          stroke-linecap="round"
+          stroke-linejoin="round"
+        >
+          <path d="M12 3v12" />
+          <path d="M7 12l5 5 5-5" />
+          <path d="M5 21h14" />
+        </svg>
+      {/if}
       Downloads
     </a>
-  </div>
+  </nav>
 </header>

dashboard/src/lib/components/ModelCard.svelte

@@ -21,7 +21,7 @@
     } | null;
     nodes?: Record<string, NodeInfo>;
     sharding?: "Pipeline" | "Tensor";
-    runtime?: "MlxRing" | "MlxIbv" | "MlxJaccl";
+    runtime?: "MlxRing" | "MlxJaccl";
     onLaunch?: () => void;
     tags?: string[];
     apiPreview?: PlacementPreview | null;
@@ -348,7 +348,7 @@
   // Debug mode state
   const isDebugMode = $derived(debugMode());
   const topology = $derived(topologyData());
-  const isRdma = $derived(runtime === "MlxIbv" || runtime === "MlxJaccl");
+  const isRdma = $derived(runtime === "MlxJaccl");
 
   // Get interface name for an IP from node data
   function getInterfaceForIp(nodeId: string, ip?: string): string | null {
@@ -567,20 +567,46 @@
     <div class="flex items-center gap-1.5 mb-2">
       <span
         class="px-1.5 py-0.5 text-xs font-mono tracking-wider uppercase bg-exo-medium-gray/30 text-exo-light-gray border border-exo-medium-gray/40"
+        title={sharding === "Pipeline"
+          ? "Pipeline: splits model into sequential stages across devices. Lower network overhead."
+          : "Tensor: splits each layer across devices. Best with high-bandwidth connections (Thunderbolt)."}
       >
         {sharding}
       </span>
       <span
         class="px-1.5 py-0.5 text-xs font-mono tracking-wider uppercase bg-exo-medium-gray/30 text-exo-light-gray border border-exo-medium-gray/40"
+        title={runtime === "MlxRing"
+          ? "Ring: standard networking. Works over any connection (Wi-Fi, Ethernet, Thunderbolt)."
+          : "RDMA: direct memory access over Thunderbolt. Significantly faster for multi-device inference."}
       >
         {runtime === "MlxRing"
           ? "MLX Ring"
-          : runtime === "MlxIbv" || runtime === "MlxJaccl"
+          : runtime === "MlxJaccl"
             ? "MLX RDMA"
             : runtime}
       </span>
     </div>
 
+    <!-- Download Status -->
+    {#if isDownloading && progress}
+      <div class="mb-2 space-y-1">
+        <div class="flex items-center justify-between text-xs font-mono">
+          <span class="text-blue-400 tracking-wider uppercase">Downloading</span
+          >
+          <span class="text-white/60"
+            >{percentage.toFixed(1)}% &middot; {formatSpeed(progress.speed)}
+            &middot; {formatEta(progress.etaMs)}</span
+          >
+        </div>
+        <div class="h-1 bg-exo-medium-gray/30 rounded overflow-hidden">
+          <div
+            class="h-full bg-blue-500/70 transition-all duration-300"
+            style="width: {percentage}%"
+          ></div>
+        </div>
+      </div>
+    {/if}
+
     <!-- Mini Topology Preview -->
     {#if placementPreview().nodes.length > 0}
       {@const preview = placementPreview()}

dashboard/src/lib/components/ModelFilterPopover.svelte

@@ -6,6 +6,7 @@
     capabilities: string[];
     sizeRange: { min: number; max: number } | null;
     downloadedOnly: boolean;
+    readyOnly: boolean;
   }
 
   type ModelFilterPopoverProps = {
@@ -190,34 +191,58 @@
       </div>
     </div>
 
-    <!-- Downloaded only -->
+    <!-- Availability filters -->
     <div>
       <h4 class="text-xs font-mono text-white/50 mb-2">Availability</h4>
-      <button
-        type="button"
-        class="px-2 py-1 text-xs font-mono rounded transition-colors {filters.downloadedOnly
-          ? 'bg-green-500/20 text-green-400 border border-green-500/30'
-          : 'bg-white/5 text-white/60 hover:bg-white/10 border border-transparent'}"
-        onclick={() =>
-          onChange({ ...filters, downloadedOnly: !filters.downloadedOnly })}
-      >
-        <svg
-          class="w-3.5 h-3.5 inline-block"
-          viewBox="0 0 24 24"
-          fill="none"
-          stroke="currentColor"
-          stroke-width="2"
-          stroke-linecap="round"
-          stroke-linejoin="round"
+      <div class="flex flex-wrap gap-1.5">
+        <button
+          type="button"
+          class="px-2 py-1 text-xs font-mono rounded transition-colors {filters.downloadedOnly
+            ? 'bg-green-500/20 text-green-400 border border-green-500/30'
+            : 'bg-white/5 text-white/60 hover:bg-white/10 border border-transparent'}"
+          onclick={() =>
+            onChange({ ...filters, downloadedOnly: !filters.downloadedOnly })}
         >
-          <path
-            class="text-white/40"
-            d="M20 20a2 2 0 0 0 2-2V8a2 2 0 0 0-2-2h-7.9a2 2 0 0 1-1.69-.9L9.6 3.9A2 2 0 0 0 7.93 3H4a2 2 0 0 0-2 2v13a2 2 0 0 0 2 2Z"
-          />
-          <path class="text-green-400" d="m9 13 2 2 4-4" />
-        </svg>
-        <span class="ml-1">Downloaded</span>
-      </button>
+          <svg
+            class="w-3.5 h-3.5 inline-block"
+            viewBox="0 0 24 24"
+            fill="none"
+            stroke="currentColor"
+            stroke-width="2"
+            stroke-linecap="round"
+            stroke-linejoin="round"
+          >
+            <path
+              class="text-white/40"
+              d="M20 20a2 2 0 0 0 2-2V8a2 2 0 0 0-2-2h-7.9a2 2 0 0 1-1.69-.9L9.6 3.9A2 2 0 0 0 7.93 3H4a2 2 0 0 0-2 2v13a2 2 0 0 0 2 2Z"
+            />
+            <path class="text-green-400" d="m9 13 2 2 4-4" />
+          </svg>
+          <span class="ml-1">Downloaded</span>
+        </button>
+        <button
+          type="button"
+          class="px-2 py-1 text-xs font-mono rounded transition-colors {filters.readyOnly
+            ? 'bg-green-500/20 text-green-400 border border-green-500/30'
+            : 'bg-white/5 text-white/60 hover:bg-white/10 border border-transparent'}"
+          onclick={() =>
+            onChange({ ...filters, readyOnly: !filters.readyOnly })}
+        >
+          <svg
+            class="w-3.5 h-3.5 inline-block"
+            viewBox="0 0 24 24"
+            fill="none"
+            stroke="currentColor"
+            stroke-width="2"
+            stroke-linecap="round"
+            stroke-linejoin="round"
+          >
+            <circle cx="12" cy="12" r="10" />
+            <path d="m9 12 2 2 4-4" />
+          </svg>
+          <span class="ml-1">Ready</span>
+        </button>
+      </div>
     </div>
 
     <!-- Size range -->

dashboard/src/lib/components/ModelPickerGroup.svelte

@@ -41,6 +41,7 @@
     onShowInfo: (group: ModelGroup) => void;
     downloadStatusMap?: Map<string, DownloadAvailability>;
     launchedAt?: number;
+    instanceStatuses?: Record<string, { status: string; statusClass: string }>;
   };
 
   let {
@@ -56,6 +57,7 @@
     onShowInfo,
     downloadStatusMap,
     launchedAt,
+    instanceStatuses = {},
   }: ModelPickerGroupProps = $props();
 
   // Group-level download status: show if any variant is downloaded
@@ -92,6 +94,12 @@
   const anyVariantFits = $derived(
     group.variants.some((v) => canModelFit(v.id)),
   );
+  // Check if any variant has an active instance (ready, loading, downloading)
+  const anyVariantHasInstance = $derived(
+    instanceStatuses
+      ? group.variants.some((v) => instanceStatuses[v.id] != null)
+      : false,
+  );
   const groupFitStatus = $derived.by((): ModelFitStatus => {
     let hasClusterCapacityOnly = false;
     for (const variant of group.variants) {
@@ -122,16 +130,35 @@
     !group.hasMultipleVariants &&
       group.variants.some((v) => v.id === selectedModelId),
   );
+
+  // Group-level instance status: show the "best" status across all variants
+  const groupInstanceStatus = $derived.by(() => {
+    if (!instanceStatuses) return null;
+    const readyStatuses = ["READY", "LOADED", "RUNNING"];
+    const loadingStatuses = ["LOADING", "WARMING UP"];
+    let bestStatus: { status: string; statusClass: string } | null = null;
+    for (const variant of group.variants) {
+      const s = instanceStatuses[variant.id];
+      if (!s) continue;
+      if (readyStatuses.includes(s.status)) return s; // Ready is best
+      if (loadingStatuses.includes(s.status) || s.status === "DOWNLOADING") {
+        bestStatus = s;
+      }
+    }
+    return bestStatus;
+  });
 </script>
 
 <div
-  class="border-b border-white/5 last:border-b-0 {!anyVariantFits
+  class="border-b border-white/5 last:border-b-0 {!anyVariantFits &&
+  !anyVariantHasInstance
     ? 'opacity-50'
     : ''}"
 >
   <!-- Main row -->
   <div
-    class="flex items-center gap-2 px-3 py-2.5 transition-colors {anyVariantFits
+    class="flex items-center gap-2 px-3 py-2.5 transition-colors {anyVariantFits ||
+    anyVariantHasInstance
       ? 'hover:bg-white/5 cursor-pointer'
       : 'cursor-not-allowed'} {isMainSelected
       ? 'bg-exo-yellow/10 border-l-2 border-exo-yellow'
@@ -141,7 +168,7 @@
         onToggleExpand();
       } else {
         const modelId = group.variants[0]?.id;
-        if (modelId && canModelFit(modelId)) {
+        if (modelId && (canModelFit(modelId) || instanceStatuses[modelId])) {
           onSelectModel(modelId);
         }
       }
@@ -155,7 +182,7 @@
           onToggleExpand();
         } else {
           const modelId = group.variants[0]?.id;
-          if (modelId && canModelFit(modelId)) {
+          if (modelId && (canModelFit(modelId) || instanceStatuses[modelId])) {
             onSelectModel(modelId);
           }
         }
@@ -346,6 +373,47 @@
       </span>
     {/if}
 
+    <!-- Instance status badge -->
+    {#if groupInstanceStatus}
+      {#if groupInstanceStatus.status === "READY" || groupInstanceStatus.status === "LOADED" || groupInstanceStatus.status === "RUNNING"}
+        <span class="flex-shrink-0" title="Running">
+          <svg
+            class="w-3 h-3 text-green-400"
+            viewBox="0 0 12 12"
+            fill="currentColor"
+          >
+            <circle cx="6" cy="6" r="5" />
+          </svg>
+        </span>
+      {:else if groupInstanceStatus.status === "DOWNLOADING"}
+        <span class="flex-shrink-0 animate-pulse" title="Downloading">
+          <svg
+            class="w-3.5 h-3.5 text-blue-400"
+            viewBox="0 0 24 24"
+            fill="none"
+            stroke="currentColor"
+            stroke-width="2"
+            stroke-linecap="round"
+            stroke-linejoin="round"
+          >
+            <path d="M21 15v4a2 2 0 01-2 2H5a2 2 0 01-2-2v-4" />
+            <polyline points="7 10 12 15 17 10" />
+            <line x1="12" y1="15" x2="12" y2="3" />
+          </svg>
+        </span>
+      {:else if groupInstanceStatus.status === "LOADING" || groupInstanceStatus.status === "WARMING UP"}
+        <span class="flex-shrink-0 animate-pulse" title="Loading">
+          <svg
+            class="w-3 h-3 text-yellow-400"
+            viewBox="0 0 12 12"
+            fill="currentColor"
+          >
+            <circle cx="6" cy="6" r="5" />
+          </svg>
+        </span>
+      {/if}
+    {/if}
+
     <!-- Check mark if selected (single-variant) -->
     {#if isMainSelected}
       <svg
@@ -420,20 +488,30 @@
       {#each group.variants as variant}
         {@const fitStatus = getModelFitStatus(variant.id)}
         {@const modelCanFit = canModelFit(variant.id)}
+        {@const variantHasInstance = instanceStatuses[variant.id] != null}
         {@const isSelected = selectedModelId === variant.id}
-        <button
-          type="button"
-          class="w-full flex items-center gap-3 px-3 py-2 pl-10 hover:bg-white/5 transition-colors text-left {!modelCanFit
+        <div
+          class="w-full flex items-center gap-3 px-3 py-2 pl-10 hover:bg-white/5 transition-colors text-left {!modelCanFit &&
+          !variantHasInstance
             ? 'opacity-50 cursor-not-allowed'
             : 'cursor-pointer'} {isSelected
             ? 'bg-exo-yellow/10 border-l-2 border-exo-yellow'
             : 'border-l-2 border-transparent'}"
-          disabled={!modelCanFit}
+          role="button"
+          tabindex="0"
           onclick={() => {
-            if (modelCanFit) {
+            if (modelCanFit || variantHasInstance) {
               onSelectModel(variant.id);
             }
           }}
+          onkeydown={(e) => {
+            if (e.key === "Enter" || e.key === " ") {
+              e.preventDefault();
+              if (modelCanFit) {
+                onSelectModel(variant.id);
+              }
+            }
+          }}
         >
           <!-- Quantization badge -->
           <span
@@ -478,6 +556,48 @@
             {/if}
           {/if}
 
+          <!-- Instance status badge -->
+          {#if instanceStatuses[variant.id]}
+            {@const instStatus = instanceStatuses[variant.id]}
+            {#if instStatus.status === "READY" || instStatus.status === "LOADED" || instStatus.status === "RUNNING"}
+              <span class="flex-shrink-0" title="Running">
+                <svg
+                  class="w-3 h-3 text-green-400"
+                  viewBox="0 0 12 12"
+                  fill="currentColor"
+                >
+                  <circle cx="6" cy="6" r="5" />
+                </svg>
+              </span>
+            {:else if instStatus.status === "DOWNLOADING"}
+              <span class="flex-shrink-0 animate-pulse" title="Downloading">
+                <svg
+                  class="w-3.5 h-3.5 text-blue-400"
+                  viewBox="0 0 24 24"
+                  fill="none"
+                  stroke="currentColor"
+                  stroke-width="2"
+                  stroke-linecap="round"
+                  stroke-linejoin="round"
+                >
+                  <path d="M21 15v4a2 2 0 01-2 2H5a2 2 0 01-2-2v-4" />
+                  <polyline points="7 10 12 15 17 10" />
+                  <line x1="12" y1="15" x2="12" y2="3" />
+                </svg>
+              </span>
+            {:else if instStatus.status === "LOADING" || instStatus.status === "WARMING UP"}
+              <span class="flex-shrink-0 animate-pulse" title="Loading">
+                <svg
+                  class="w-3 h-3 text-yellow-400"
+                  viewBox="0 0 12 12"
+                  fill="currentColor"
+                >
+                  <circle cx="6" cy="6" r="5" />
+                </svg>
+              </span>
+            {/if}
+          {/if}
+
           <!-- Check mark if selected -->
           {#if isSelected}
             <svg
@@ -490,7 +610,36 @@
               />
             </svg>
           {/if}
-        </button>
+
+          <!-- Info button -->
+          <button
+            type="button"
+            class="p-1 rounded hover:bg-white/10 transition-colors flex-shrink-0"
+            onclick={(e) => {
+              e.stopPropagation();
+              onShowInfo({
+                id: variant.id,
+                name: variant.name || variant.id,
+                capabilities: group.capabilities,
+                family: group.family,
+                variants: [variant],
+                smallestVariant: variant,
+                hasMultipleVariants: false,
+              });
+            }}
+            title="View variant details"
+          >
+            <svg
+              class="w-4 h-4 text-white/30 hover:text-white/50"
+              viewBox="0 0 24 24"
+              fill="currentColor"
+            >
+              <path
+                d="M12 2C6.48 2 2 6.48 2 12s4.48 10 10 10 10-4.48 10-10S17.52 2 12 2zm1 15h-2v-6h2v6zm0-8h-2V7h2v2z"
+              />
+            </svg>
+          </button>
+        </div>
       {/each}
     </div>
   {/if}

dashboard/src/lib/components/ModelPickerModal.svelte

@@ -36,6 +36,7 @@
     capabilities: string[];
     sizeRange: { min: number; max: number } | null;
     downloadedOnly: boolean;
+    readyOnly: boolean;
   }
 
   interface HuggingFaceModel {
@@ -49,6 +50,11 @@
 
   type ModelFitStatus = "fits_now" | "fits_cluster_capacity" | "too_large";
 
+  export type InstanceStatus = {
+    status: string;
+    statusClass: string;
+  };
+
   type ModelPickerModalProps = {
     isOpen: boolean;
     models: ModelInfo[];
@@ -75,6 +81,7 @@
         macmon_info?: { memory?: { ram_total?: number } };
       }
     >;
+    instanceStatuses?: Record<string, InstanceStatus>;
   };
 
   let {
@@ -96,6 +103,7 @@
     usedMemoryGB,
     downloadsData,
     topologyNodes,
+    instanceStatuses = {},
   }: ModelPickerModalProps = $props();
 
   // Local state
@@ -107,6 +115,7 @@
     capabilities: [],
     sizeRange: null,
     downloadedOnly: false,
+    readyOnly: false,
   });
   let infoGroup = $state<ModelGroup | null>(null);
 
@@ -440,6 +449,16 @@
       );
     }
 
+    // Filter to ready/running models only
+    if (filters.readyOnly) {
+      result = result.filter((g) =>
+        g.variants.some((v) => {
+          const s = instanceStatuses[v.id];
+          return s && s.statusClass === "ready";
+        }),
+      );
+    }
+
     // Sort: fits-now first, then fits-cluster-capacity, then too-large
     result.sort((a, b) => {
       const getGroupFitRank = (group: ModelGroup): number => {
@@ -512,6 +531,18 @@
     );
   });
 
+  // Split filtered groups into recommended (fits_now) and others for visual separation
+  const recommendedGroups = $derived(
+    filteredGroups.filter((g) =>
+      g.variants.some((v) => getModelFitStatus(v.id) === "fits_now"),
+    ),
+  );
+  const otherGroups = $derived(
+    filteredGroups.filter(
+      (g) => !g.variants.some((v) => getModelFitStatus(v.id) === "fits_now"),
+    ),
+  );
+
   function toggleGroupExpanded(groupId: string) {
     const next = new Set(expandedGroups);
     if (next.has(groupId)) {
@@ -538,13 +569,19 @@
   }
 
   function clearFilters() {
-    filters = { capabilities: [], sizeRange: null, downloadedOnly: false };
+    filters = {
+      capabilities: [],
+      sizeRange: null,
+      downloadedOnly: false,
+      readyOnly: false,
+    };
   }
 
   const hasActiveFilters = $derived(
     filters.capabilities.length > 0 ||
       filters.sizeRange !== null ||
-      filters.downloadedOnly,
+      filters.downloadedOnly ||
+      filters.readyOnly,
   );
 </script>
 
@@ -813,6 +850,7 @@
                 onShowInfo={(g) => (infoGroup = g)}
                 downloadStatusMap={getVariantDownloadMap(group)}
                 launchedAt={recentTimestamps.get(group.variants[0]?.id ?? "")}
+                {instanceStatuses}
               />
             {/each}
           {/if}
@@ -840,7 +878,34 @@
             {/if}
           </div>
         {:else}
-          {#each filteredGroups as group}
+          <!-- Recommended for your cluster -->
+          {#if recommendedGroups.length > 0 && otherGroups.length > 0 && !searchQuery.trim()}
+            <div
+              class="sticky top-0 z-10 flex items-center gap-2 px-3 py-2 bg-green-950/60 border-b border-green-500/20 backdrop-blur-sm"
+            >
+              <svg
+                class="w-3.5 h-3.5 text-green-400 flex-shrink-0"
+                fill="none"
+                viewBox="0 0 24 24"
+                stroke="currentColor"
+                stroke-width="2"
+              >
+                <path
+                  stroke-linecap="round"
+                  stroke-linejoin="round"
+                  d="M9 12.75L11.25 15 15 9.75M21 12a9 9 0 11-18 0 9 9 0 0118 0z"
+                />
+              </svg>
+              <span
+                class="text-xs font-mono text-green-400 tracking-wider uppercase"
+                >Recommended for your cluster</span
+              >
+              <span class="text-xs font-mono text-green-400/50"
+                >— fits in available memory</span
+              >
+            </div>
+          {/if}
+          {#each recommendedGroups as group}
             <ModelPickerGroup
               {group}
               isExpanded={expandedGroups.has(group.id)}
@@ -853,6 +918,34 @@
               {onToggleFavorite}
               onShowInfo={(g) => (infoGroup = g)}
               downloadStatusMap={getVariantDownloadMap(group)}
+              {instanceStatuses}
+            />
+          {/each}
+          <!-- Other models -->
+          {#if otherGroups.length > 0 && recommendedGroups.length > 0 && !searchQuery.trim()}
+            <div
+              class="sticky top-0 z-10 flex items-center gap-2 px-3 py-2 bg-exo-dark-gray/80 border-y border-exo-medium-gray/20 backdrop-blur-sm"
+            >
+              <span
+                class="text-xs font-mono text-white/40 tracking-wider uppercase"
+                >Other models</span
+              >
+            </div>
+          {/if}
+          {#each otherGroups as group}
+            <ModelPickerGroup
+              {group}
+              isExpanded={expandedGroups.has(group.id)}
+              isFavorite={favorites.has(group.id)}
+              {selectedModelId}
+              {canModelFit}
+              {getModelFitStatus}
+              onToggleExpand={() => toggleGroupExpanded(group.id)}
+              onSelectModel={handleSelect}
+              {onToggleFavorite}
+              onShowInfo={(g) => (infoGroup = g)}
+              downloadStatusMap={getVariantDownloadMap(group)}
+              {instanceStatuses}
             />
           {/each}
         {/if}
@@ -875,6 +968,11 @@
             >Downloaded</span
           >
         {/if}
+        {#if filters.readyOnly}
+          <span class="px-1.5 py-0.5 bg-green-500/20 text-green-400 rounded"
+            >Ready</span
+          >
+        {/if}
         {#if filters.sizeRange}
           <span class="px-1.5 py-0.5 bg-exo-yellow/20 text-exo-yellow rounded">
             {filters.sizeRange.min}GB - {filters.sizeRange.max}GB

dashboard/src/lib/components/ToastContainer.svelte

@@ -0,0 +1,117 @@
+<script lang="ts">
+  import { toasts, dismissToast, type Toast } from "$lib/stores/toast.svelte";
+  import { fly, fade } from "svelte/transition";
+  import { flip } from "svelte/animate";
+
+  const items = $derived(toasts());
+
+  const typeStyles: Record<
+    Toast["type"],
+    { border: string; icon: string; iconColor: string }
+  > = {
+    success: {
+      border: "border-l-green-500",
+      icon: "M9 12.75L11.25 15 15 9.75M21 12a9 9 0 11-18 0 9 9 0 0118 0z",
+      iconColor: "text-green-400",
+    },
+    error: {
+      border: "border-l-red-500",
+      icon: "M12 9v3.75m9-.75a9 9 0 11-18 0 9 9 0 0118 0zm-9 3.75h.008v.008H12v-.008z",
+      iconColor: "text-red-400",
+    },
+    warning: {
+      border: "border-l-yellow-500",
+      icon: "M12 9v3.75m-9.303 3.376c-.866 1.5.217 3.374 1.948 3.374h14.71c1.73 0 2.813-1.874 1.948-3.374L13.949 3.378c-.866-1.5-3.032-1.5-3.898 0L2.697 16.126z",
+      iconColor: "text-yellow-400",
+    },
+    info: {
+      border: "border-l-blue-500",
+      icon: "M11.25 11.25l.041-.02a.75.75 0 011.063.852l-.708 2.836a.75.75 0 001.063.853l.041-.021M21 12a9 9 0 11-18 0 9 9 0 0118 0zm-9-3.75h.008v.008H12V8.25z",
+      iconColor: "text-blue-400",
+    },
+  };
+</script>
+
+{#if items.length > 0}
+  <div
+    class="fixed bottom-6 right-6 z-[9999] flex flex-col gap-2 pointer-events-none"
+    role="log"
+    aria-live="polite"
+    aria-label="Notifications"
+  >
+    {#each items as toast (toast.id)}
+      {@const style = typeStyles[toast.type]}
+      <div
+        class="pointer-events-auto max-w-sm w-80 bg-exo-dark-gray/95 backdrop-blur-sm border border-exo-medium-gray/60 border-l-[3px] {style.border} rounded shadow-lg shadow-black/40"
+        in:fly={{ x: 80, duration: 250 }}
+        out:fade={{ duration: 150 }}
+        animate:flip={{ duration: 200 }}
+        role="alert"
+      >
+        <div class="flex items-start gap-3 px-4 py-3">
+          <!-- Icon -->
+          <svg
+            class="w-5 h-5 flex-shrink-0 mt-0.5 {style.iconColor}"
+            fill="none"
+            viewBox="0 0 24 24"
+            stroke-width="1.5"
+            stroke="currentColor"
+          >
+            <path
+              stroke-linecap="round"
+              stroke-linejoin="round"
+              d={style.icon}
+            />
+          </svg>
+
+          <!-- Message -->
+          <p class="flex-1 text-sm text-white/90 font-mono leading-snug">
+            {toast.message}
+          </p>
+
+          <!-- Dismiss button -->
+          <button
+            onclick={() => dismissToast(toast.id)}
+            class="flex-shrink-0 p-0.5 text-white/40 hover:text-white/80 transition-colors cursor-pointer"
+            aria-label="Dismiss notification"
+          >
+            <svg
+              class="w-4 h-4"
+              fill="none"
+              viewBox="0 0 24 24"
+              stroke-width="2"
+              stroke="currentColor"
+            >
+              <path
+                stroke-linecap="round"
+                stroke-linejoin="round"
+                d="M6 18L18 6M6 6l12 12"
+              />
+            </svg>
+          </button>
+        </div>
+
+        <!-- Auto-dismiss progress bar -->
+        {#if toast.duration > 0}
+          <div class="h-0.5 bg-white/5 rounded-b overflow-hidden">
+            <div
+              class="h-full {style.border.replace('border-l-', 'bg-')}/60"
+              style="animation: shrink {toast.duration}ms linear forwards"
+            ></div>
+          </div>
+        {/if}
+      </div>
+    {/each}
+  </div>
+{/if}
+
+<style>
+  @keyframes shrink {
+    from {
+      width: 100%;
+    }
+    to {
+      width: 0%;
+    }
+  }
+</style>

dashboard/src/lib/components/index.ts

@@ -12,3 +12,4 @@ export { default as HuggingFaceResultItem } from "./HuggingFaceResultItem.svelte
 export { default as ModelFilterPopover } from "./ModelFilterPopover.svelte";
 export { default as ModelPickerGroup } from "./ModelPickerGroup.svelte";
 export { default as ModelPickerModal } from "./ModelPickerModal.svelte";
+export { default as ChatModelSelector } from "./ChatModelSelector.svelte";

dashboard/src/lib/stores/app.svelte.ts

@@ -168,7 +168,7 @@ export interface ModelDownloadStatus {
 export interface PlacementPreview {
   model_id: string;
   sharding: "Pipeline" | "Tensor";
-  instance_meta: "MlxRing" | "MlxIbv" | "MlxJaccl";
+  instance_meta: "MlxRing" | "MlxJaccl";
   instance: unknown | null;
   memory_delta_by_node: Record<string, number> | null;
   error: string | null;
@@ -219,7 +219,6 @@ interface RawStateResponse {
     string,
     {
       MlxRingInstance?: Instance;
-      MlxIbvInstance?: Instance;
       MlxJacclInstance?: Instance;
     }
   >;
@@ -590,6 +589,12 @@ class AppStore {
   // Image editing state
   editingImage = $state<EditingImage | null>(null);
 
+  /** True when the backend is reachable. */
+  isConnected = $state<boolean>(true);
+  /** Number of consecutive fetch failures. */
+  private consecutiveFailures = 0;
+  private static readonly CONNECTION_LOST_THRESHOLD = 3;
+
   private fetchInterval: ReturnType<typeof setInterval> | null = null;
   private previewsInterval: ReturnType<typeof setInterval> | null = null;
   private lastConversationPersistTs = 0;
@@ -842,6 +847,10 @@ class AppStore {
     this.thinkingEnabled = conversation.enableThinking ?? true;
     this.refreshConversationModelFromInstances();
 
+    // Sync global selection to the loaded conversation's model so reactive
+    // effects in +page.svelte can determine the correct chat launch state.
+    this.selectedChatModel = conversation.modelId || "";
+
     return true;
   }
 
@@ -912,11 +921,7 @@ class AppStore {
 
     let instanceType: string | null = null;
     if (instanceTag === "MlxRingInstance") instanceType = "MLX Ring";
-    else if (
-      instanceTag === "MlxIbvInstance" ||
-      instanceTag === "MlxJacclInstance"
-    )
-      instanceType = "MLX RDMA";
+    else if (instanceTag === "MlxJacclInstance") instanceType = "MLX RDMA";
 
     let sharding: string | null = null;
     const inst = instance as {
@@ -1295,7 +1300,19 @@ class AppStore {
       // Thunderbolt bridge status per node
       this.nodeThunderboltBridge = data.nodeThunderboltBridge ?? {};
       this.lastUpdate = Date.now();
+      // Connection recovered
+      if (!this.isConnected) {
+        this.isConnected = true;
+      }
+      this.consecutiveFailures = 0;
     } catch (error) {
+      this.consecutiveFailures++;
+      if (
+        this.consecutiveFailures >= AppStore.CONNECTION_LOST_THRESHOLD &&
+        this.isConnected
+      ) {
+        this.isConnected = false;
+      }
       console.error("Error fetching state:", error);
     }
   }
@@ -1836,7 +1853,7 @@ class AppStore {
           assistantMessage.id,
           (msg) => {
             msg.content =
-              "Error: No model available. Please launch an instance first.";
+              "No model is loaded yet. Select a model from the sidebar to get started — it will download and load automatically.";
           },
         );
         this.syncActiveMessagesIfNeeded(targetConversationId);
@@ -2307,7 +2324,7 @@ class AppStore {
       const modelToUse = this.getModelForRequest();
       if (!modelToUse) {
         throw new Error(
-          "No model selected and no running instances available. Please launch an instance first.",
+          "No model is loaded yet. Select a model from the sidebar to get started — it will download and load automatically.",
         );
       }
 
@@ -3250,6 +3267,9 @@ export const setChatSidebarVisible = (visible: boolean) =>
   appStore.setChatSidebarVisible(visible);
 export const refreshState = () => appStore.fetchState();
 
+// Connection status
+export const isConnected = () => appStore.isConnected;
+
 // Node identities (for OS version mismatch detection)
 export const nodeIdentities = () => appStore.nodeIdentities;
 

dashboard/src/lib/stores/toast.svelte.ts

@@ -0,0 +1,87 @@
+/**
+ * Toast notification store - Global notification system for the EXO dashboard.
+ *
+ * Usage:
+ *   import { addToast, dismissToast, toasts } from "$lib/stores/toast.svelte";
+ *   addToast({ type: "success", message: "Model launched" });
+ *   addToast({ type: "error", message: "Connection lost", persistent: true });
+ */
+
+type ToastType = "success" | "error" | "warning" | "info";
+
+export interface Toast {
+  id: string;
+  type: ToastType;
+  message: string;
+  /** Auto-dismiss after this many ms. 0 = persistent (must be dismissed manually). */
+  duration: number;
+  createdAt: number;
+}
+
+interface ToastInput {
+  type: ToastType;
+  message: string;
+  /** If true, toast stays until manually dismissed. Default: false. */
+  persistent?: boolean;
+  /** Auto-dismiss duration in ms. Default: 4000 for success/info, 6000 for error/warning. */
+  duration?: number;
+}
+
+const DEFAULT_DURATIONS: Record<ToastType, number> = {
+  success: 4000,
+  info: 4000,
+  warning: 6000,
+  error: 6000,
+};
+
+let toastList = $state<Toast[]>([]);
+const timers = new Map<string, ReturnType<typeof setTimeout>>();
+
+function generateId(): string {
+  return `toast-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`;
+}
+
+export function addToast(input: ToastInput): string {
+  const id = generateId();
+  const duration = input.persistent
+    ? 0
+    : (input.duration ?? DEFAULT_DURATIONS[input.type]);
+
+  const toast: Toast = {
+    id,
+    type: input.type,
+    message: input.message,
+    duration,
+    createdAt: Date.now(),
+  };
+
+  toastList = [...toastList, toast];
+
+  if (duration > 0) {
+    const timer = setTimeout(() => dismissToast(id), duration);
+    timers.set(id, timer);
+  }
+
+  return id;
+}
+
+export function dismissToast(id: string): void {
+  const timer = timers.get(id);
+  if (timer) {
+    clearTimeout(timer);
+    timers.delete(id);
+  }
+  toastList = toastList.filter((t) => t.id !== id);
+}
+
+/** Dismiss all toasts matching a message (useful for dedup). */
+export function dismissByMessage(message: string): void {
+  const matching = toastList.filter((t) => t.message === message);
+  for (const t of matching) {
+    dismissToast(t.id);
+  }
+}
+
+export function toasts(): Toast[] {
+  return toastList;
+}

dashboard/src/routes/+layout.svelte

@@ -1,5 +1,7 @@
 <script lang="ts">
   import "../app.css";
+  import ToastContainer from "$lib/components/ToastContainer.svelte";
+  import ConnectionBanner from "$lib/components/ConnectionBanner.svelte";
 
   let { children } = $props();
 </script>
@@ -10,5 +12,7 @@
 </svelte:head>
 
 <div class="min-h-screen bg-background text-foreground">
+  <ConnectionBanner />
   {@render children?.()}
+  <ToastContainer />
 </div>

dashboard/src/routes/downloads/+page.svelte

@@ -412,7 +412,7 @@
                   <div>{col.label}</div>
                   {#if col.diskAvailable != null}
                     <div
-                      class="text-[9px] text-exo-light-gray/60 normal-case tracking-normal mt-0.5"
+                      class="text-[9px] text-white/70 normal-case tracking-normal mt-0.5"
                     >
                       {formatBytes(col.diskAvailable)} free
                     </div>
@@ -436,7 +436,7 @@
                       </div>
                       {#if row.prettyName}
                         <div
-                          class="text-[10px] text-exo-light-gray/60"
+                          class="text-[10px] text-white/60"
                           title={row.modelId}
                         >
                           {row.modelId}
@@ -450,7 +450,7 @@
                       title="View model details"
                     >
                       <svg
-                        class="w-4 h-4 text-white/30 hover:text-white/60"
+                        class="w-4 h-4 text-white/60 hover:text-white/80"
                         viewBox="0 0 24 24"
                         fill="currentColor"
                       >
@@ -469,11 +469,11 @@
                   <td class="px-4 py-3 text-center align-middle">
                     {#if cell.kind === "completed"}
                       <div
-                        class="flex flex-col items-center gap-0.5"
+                        class="flex flex-col items-center gap-1"
                         title="Completed ({formatBytes(cell.totalBytes)})"
                       >
                         <svg
-                          class="w-5 h-5 text-green-400"
+                          class="w-7 h-7 text-green-400"
                           viewBox="0 0 20 20"
                           fill="currentColor"
                         >
@@ -483,18 +483,18 @@
                             clip-rule="evenodd"
                           ></path>
                         </svg>
-                        <span class="text-[10px] text-exo-light-gray/70"
+                        <span class="text-xs text-white/70"
                           >{formatBytes(cell.totalBytes)}</span
                         >
                         <button
                           type="button"
-                          class="text-exo-light-gray/40 hover:text-red-400 transition-colors mt-0.5"
+                          class="text-white/50 hover:text-red-400 transition-colors mt-0.5 cursor-pointer"
                           onclick={() =>
                             deleteDownload(col.nodeId, row.modelId)}
                           title="Delete from this node"
                         >
                           <svg
-                            class="w-3.5 h-3.5"
+                            class="w-5 h-5"
                             viewBox="0 0 20 20"
                             fill="none"
                             stroke="currentColor"
@@ -517,11 +517,11 @@
                           cell.speed,
                         )} - ETA {formatEta(cell.etaMs)}"
                       >
-                        <span class="text-exo-yellow text-xs font-medium"
+                        <span class="text-exo-yellow text-sm font-medium"
                           >{clampPercent(cell.percentage).toFixed(1)}%</span
                         >
                         <div
-                          class="w-14 h-1.5 bg-exo-black/60 rounded-sm overflow-hidden"
+                          class="w-16 h-2 bg-exo-black/60 rounded-sm overflow-hidden"
                         >
                           <div
                             class="h-full bg-gradient-to-r from-exo-yellow to-exo-yellow/70 transition-all duration-300"
@@ -530,25 +530,25 @@
                             ).toFixed(1)}%"
                           ></div>
                         </div>
-                        <span class="text-[9px] text-exo-light-gray/60"
+                        <span class="text-[10px] text-white/70"
                           >{formatSpeed(cell.speed)}</span
                         >
                       </div>
                     {:else if cell.kind === "pending"}
                       <div
-                        class="flex flex-col items-center gap-0.5"
+                        class="flex flex-col items-center gap-1"
                         title={cell.downloaded > 0
-                          ? `${formatBytes(cell.downloaded)} / ${formatBytes(cell.total)} downloaded`
+                          ? `${formatBytes(cell.downloaded)} / ${formatBytes(cell.total)} downloaded (paused)`
                           : "Download pending"}
                       >
                         {#if cell.downloaded > 0 && cell.total > 0}
-                          <span class="text-exo-light-gray/70 text-[10px]"
+                          <span class="text-white/70 text-xs"
                             >{formatBytes(cell.downloaded)} / {formatBytes(
                               cell.total,
                             )}</span
                           >
                           <div
-                            class="w-full h-1 bg-white/10 rounded-full overflow-hidden"
+                            class="w-full h-1.5 bg-white/10 rounded-full overflow-hidden"
                           >
                             <div
                               class="h-full bg-exo-light-gray/40 rounded-full"
@@ -558,21 +558,65 @@
                               ).toFixed(1)}%"
                             ></div>
                           </div>
-                          <span class="text-exo-light-gray/40 text-[9px]"
-                            >paused</span
+                          {#if row.shardMetadata}
+                            <button
+                              type="button"
+                              class="text-white/50 hover:text-exo-yellow transition-colors cursor-pointer"
+                              onclick={() =>
+                                startDownload(col.nodeId, row.shardMetadata!)}
+                              title="Resume download on this node"
+                            >
+                              <svg
+                                class="w-5 h-5"
+                                viewBox="0 0 20 20"
+                                fill="none"
+                                stroke="currentColor"
+                                stroke-width="2"
+                              >
+                                <path
+                                  d="M10 3v10m0 0l-3-3m3 3l3-3M3 17h14"
+                                  stroke-linecap="round"
+                                  stroke-linejoin="round"
+                                ></path>
+                              </svg>
+                            </button>
+                          {:else}
+                            <span class="text-white/50 text-[10px]">paused</span
+                            >
+                          {/if}
+                        {:else if row.shardMetadata}
+                          <button
+                            type="button"
+                            class="text-white/50 hover:text-exo-yellow transition-colors cursor-pointer"
+                            onclick={() =>
+                              startDownload(col.nodeId, row.shardMetadata!)}
+                            title="Start download on this node"
                           >
+                            <svg
+                              class="w-6 h-6"
+                              viewBox="0 0 20 20"
+                              fill="none"
+                              stroke="currentColor"
+                              stroke-width="2"
+                            >
+                              <path
+                                d="M10 3v10m0 0l-3-3m3 3l3-3M3 17h14"
+                                stroke-linecap="round"
+                                stroke-linejoin="round"
+                              ></path>
+                            </svg>
+                          </button>
                         {:else}
-                          <span class="text-exo-light-gray/50 text-sm">...</span
-                          >
+                          <span class="text-white/40 text-sm">...</span>
                         {/if}
                       </div>
                     {:else if cell.kind === "failed"}
                       <div
-                        class="flex flex-col items-center gap-0.5"
+                        class="flex flex-col items-center gap-1"
                         title="Download failed"
                       >
                         <svg
-                          class="w-5 h-5 text-red-400"
+                          class="w-7 h-7 text-red-400"
                           viewBox="0 0 20 20"
                           fill="currentColor"
                         >
@@ -585,13 +629,13 @@
                         {#if row.shardMetadata}
                           <button
                             type="button"
-                            class="text-exo-light-gray/40 hover:text-exo-yellow transition-colors"
+                            class="text-white/50 hover:text-exo-yellow transition-colors cursor-pointer"
                             onclick={() =>
                               startDownload(col.nodeId, row.shardMetadata!)}
                             title="Retry download on this node"
                           >
                             <svg
-                              class="w-3.5 h-3.5"
+                              class="w-5 h-5"
                               viewBox="0 0 20 20"
                               fill="none"
                               stroke="currentColor"
@@ -617,13 +661,13 @@
                         {#if row.shardMetadata}
                           <button
                             type="button"
-                            class="text-exo-light-gray/30 hover:text-exo-yellow transition-colors mt-0.5 opacity-0 group-hover:opacity-100"
+                            class="text-white/50 hover:text-exo-yellow transition-colors mt-0.5 opacity-0 group-hover:opacity-100 cursor-pointer"
                             onclick={() =>
                               startDownload(col.nodeId, row.shardMetadata!)}
                             title="Download to this node"
                           >
                             <svg
-                              class="w-3.5 h-3.5"
+                              class="w-5 h-5"
                               viewBox="0 0 20 20"
                               fill="none"
                               stroke="currentColor"

nix/mlx.nix

@@ -41,16 +41,16 @@ let
 
   mlx = stdenv.mkDerivation rec {
     pname = "mlx";
-    version = let v = "0.30.7.dev20260220+db176851"; in
+    version = let v = "0.30.7.dev20260224+e862b122"; in
       assert v == uvLockMlxVersion || throw "MLX version mismatch: nix/mlx.nix has ${v} but uv.lock has ${uvLockMlxVersion}. Update both the version and hash in nix/mlx.nix.";
       v;
     pyproject = true;
 
     src = fetchFromGitHub {
-      owner = "exo-explore";
-      repo = "mlx";
-      rev = "db176851b031631b414890d6bd31d3dfb2e82d25";
-      hash = "sha256-/76lKDMmAjs7Qqh9V0oIBoNPq1kv819JZ+UbVFF8pd4=";
+      owner = "rltakashige";
+      repo = "mlx-jaccl-fix-small-recv";
+      rev = "e862b1223a2310d4cc8df1135aed42f5246bc50a";
+      hash = "sha256-GosFIWxIB48Egb1MqJrR3xhsUsQeWdRk5rV93USY6wQ=";
     };
 
     patches = [

packaging/dmg/create-dmg.sh

@@ -0,0 +1,112 @@
+#!/usr/bin/env bash
+# create-dmg.sh — Build a polished macOS DMG installer for EXO
+#
+# Usage:
+#   ./packaging/dmg/create-dmg.sh <app-path> <output-dmg> [volume-name]
+#
+# Example:
+#   ./packaging/dmg/create-dmg.sh output/EXO.app EXO-1.0.0.dmg "EXO"
+#
+# Creates a DMG with:
+#   - Custom background image with drag-to-Applications arrow
+#   - App icon on left, Applications alias on right
+#   - Proper window size and icon positioning
+set -euo pipefail
+
+APP_PATH="${1:?Usage: create-dmg.sh <app-path> <output-dmg> [volume-name]}"
+OUTPUT_DMG="${2:?Usage: create-dmg.sh <app-path> <output-dmg> [volume-name]}"
+VOLUME_NAME="${3:-EXO}"
+
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+BACKGROUND_SCRIPT="${SCRIPT_DIR}/generate-background.py"
+TEMP_DIR="$(mktemp -d)"
+DMG_STAGING="${TEMP_DIR}/dmg-root"
+TEMP_DMG="${TEMP_DIR}/temp.dmg"
+BACKGROUND_PNG="${TEMP_DIR}/background.png"
+
+cleanup() { rm -rf "$TEMP_DIR"; }
+trap cleanup EXIT
+
+echo "==> Creating DMG installer for ${VOLUME_NAME}"
+
+# ── Step 1: Generate background image ────────────────────────────────────────
+if command -v python3 &>/dev/null; then
+  python3 "$BACKGROUND_SCRIPT" "$BACKGROUND_PNG"
+  echo "    Background image generated"
+else
+  echo "    Warning: python3 not found, skipping custom background"
+  BACKGROUND_PNG=""
+fi
+
+# ── Step 2: Prepare staging directory ─────────────────────────────────────────
+mkdir -p "$DMG_STAGING"
+cp -R "$APP_PATH" "$DMG_STAGING/"
+ln -s /Applications "$DMG_STAGING/Applications"
+
+# ── Step 3: Create writable DMG ──────────────────────────────────────────────
+# Calculate required size (app size + 20MB headroom)
+APP_SIZE_KB=$(du -sk "$APP_PATH" | cut -f1)
+DMG_SIZE_KB=$((APP_SIZE_KB + 20480))
+
+hdiutil create \
+  -volname "$VOLUME_NAME" \
+  -size "${DMG_SIZE_KB}k" \
+  -fs HFS+ \
+  -layout SPUD \
+  "$TEMP_DMG"
+
+# ── Step 4: Mount and configure ──────────────────────────────────────────────
+MOUNT_DIR=$(hdiutil attach "$TEMP_DMG" -readwrite -noverify | awk -F'\t' '/Apple_HFS/ {gsub(/^[[:space:]]+|[[:space:]]+$/, "", $NF); print $NF}')
+echo "    Mounted at: $MOUNT_DIR"
+
+# Copy contents
+cp -R "$DMG_STAGING/"* "$MOUNT_DIR/"
+
+# Add background image
+if [[ -n $BACKGROUND_PNG && -f $BACKGROUND_PNG ]]; then
+  mkdir -p "$MOUNT_DIR/.background"
+  cp "$BACKGROUND_PNG" "$MOUNT_DIR/.background/background.png"
+fi
+
+# ── Step 5: Configure window appearance via AppleScript ──────────────────────
+# Window: 800×400, app icon on left, Applications on right (matches Ollama layout)
+# Background image is 1600×740 (2× retina for 800×400 logical window).
+APP_NAME="$(basename "$APP_PATH")"
+
+osascript <<APPLESCRIPT
+tell application "Finder"
+    tell disk "$VOLUME_NAME"
+        open
+        set current view of container window to icon view
+        set toolbar visible of container window to false
+        set statusbar visible of container window to false
+        set bounds of container window to {200, 120, 1000, 520}
+        set opts to icon view options of container window
+        set icon size of opts to 128
+        set text size of opts to 12
+        set arrangement of opts to not arranged
+        if exists file ".background:background.png" then
+            set background picture of opts to file ".background:background.png"
+        end if
+        set position of item "$APP_NAME" of container window to {200, 190}
+        set position of item "Applications" of container window to {600, 190}
+        close
+        open
+        update without registering applications
+        delay 1
+        close
+    end tell
+end tell
+APPLESCRIPT
+
+echo "    Window layout configured"
+
+# Ensure Finder updates are flushed
+sync
+
+# ── Step 6: Finalise ─────────────────────────────────────────────────────────
+hdiutil detach "$MOUNT_DIR" -quiet
+hdiutil convert "$TEMP_DMG" -format UDZO -imagekey zlib-level=9 -o "$OUTPUT_DMG"
+
+echo "==> DMG created: $OUTPUT_DMG"
+echo "    Size: $(du -h "$OUTPUT_DMG" | cut -f1)"

packaging/dmg/generate-background.py

@@ -0,0 +1,91 @@
+#!/usr/bin/env python3
+"""Generate the DMG background image with a centered drag-to-Applications arrow.
+
+The output is a 1600×740 retina PNG (2× for 800×400 logical window).
+Icons are positioned at (200, 190) and (600, 190) in logical coordinates;
+the arrow is drawn centered between them.
+
+Usage:
+    python3 generate-background.py [output.png]
+
+If no output path is given, overwrites the bundled background.png in-place.
+"""
+
+from __future__ import annotations
+
+import math
+import sys
+from pathlib import Path
+
+from PIL import Image, ImageDraw
+
+# Retina dimensions (2× logical 800×400)
+WIDTH = 1600
+HEIGHT = 740
+
+# Icon positions in logical coords → retina coords
+# App icon at (200, 190), Applications at (600, 190)
+APP_X = 200 * 2  # 400
+APPS_X = 600 * 2  # 1200
+ICON_Y = 190 * 2  # 380
+
+# Arrow drawn between icons, slightly above icon center
+ARROW_START_X = APP_X + 160  # past the icon
+ARROW_END_X = APPS_X - 160  # before the Applications icon
+ARROW_Y = ICON_Y  # same height as icons
+ARROW_RISE = 120  # upward arc height
+
+
+def draw_arrow(draw: ImageDraw.ImageDraw) -> None:
+    """Draw a hand-drawn-style curved arrow from app icon toward Applications."""
+    color = (30, 30, 30)
+    line_width = 8
+
+    # Compute bezier curve points for a gentle upward arc
+    points: list[tuple[float, float]] = []
+    steps = 80
+    for i in range(steps + 1):
+        t = i / steps
+        # Quadratic bezier: start → control → end
+        cx = (ARROW_START_X + ARROW_END_X) / 2
+        cy = ARROW_Y - ARROW_RISE
+        x = (1 - t) ** 2 * ARROW_START_X + 2 * (1 - t) * t * cx + t**2 * ARROW_END_X
+        y = (1 - t) ** 2 * ARROW_Y + 2 * (1 - t) * t * cy + t**2 * ARROW_Y
+        points.append((x, y))
+
+    # Draw the curve as connected line segments
+    for i in range(len(points) - 1):
+        draw.line([points[i], points[i + 1]], fill=color, width=line_width)
+
+    # Arrowhead at the end
+    end_x, end_y = points[-1]
+    # Direction from second-to-last to last point
+    prev_x, prev_y = points[-3]
+    angle = math.atan2(end_y - prev_y, end_x - prev_x)
+    head_len = 36
+    head_angle = math.radians(25)
+
+    left_x = end_x - head_len * math.cos(angle - head_angle)
+    left_y = end_y - head_len * math.sin(angle - head_angle)
+    right_x = end_x - head_len * math.cos(angle + head_angle)
+    right_y = end_y - head_len * math.sin(angle + head_angle)
+
+    draw.polygon(
+        [(end_x, end_y), (left_x, left_y), (right_x, right_y)],
+        fill=color,
+    )
+
+
+def generate_background(output_path: str) -> None:
+    """Generate a white DMG background with a centered arrow."""
+    img = Image.new("RGBA", (WIDTH, HEIGHT), (255, 255, 255, 255))
+    draw = ImageDraw.Draw(img)
+    draw_arrow(draw)
+    img.save(output_path, "PNG")
+
+
+if __name__ == "__main__":
+    default_output = str(Path(__file__).parent / "background.png")
+    out = sys.argv[1] if len(sys.argv) >= 2 else default_output
+    generate_background(out)
+    print(f"Background image written to {out}")

pyproject.toml

@@ -64,7 +64,7 @@ members = [
 
 [tool.uv.sources]
 exo_pyo3_bindings = { workspace = true }
-mlx = { git = "https://github.com/exo-explore/mlx.git", branch = "pipe-sidechannel", marker = "sys_platform == 'darwin'" }
+mlx = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git", branch = "address-rdma-gpu-locks", marker = "sys_platform == 'darwin'" }
 #mlx-lm = { git = "https://github.com/davidmcc73/mlx-lm", branch = "stable" }
 # Uncomment to use local mlx/mlx-lm development versions:
 # mlx = { path = "/Users/Shared/mlx", editable=true }

rust/exo_pyo3_bindings/exo_pyo3_bindings.pyi

@@ -104,6 +104,12 @@ class NetworkingHandle:
         will sleep until a message is sent.
         """
 
+@typing.final
+class MessageTooLargeError(builtins.Exception):
+    def __new__(cls, *args: typing.Any) -> MessageTooLargeError: ...
+    def __repr__(self) -> builtins.str: ...
+    def __str__(self) -> builtins.str: ...
+
 @typing.final
 class NoPeersSubscribedToTopicError(builtins.Exception):
     def __new__(cls, *args: typing.Any) -> NoPeersSubscribedToTopicError: ...

rust/exo_pyo3_bindings/src/networking.rs

@@ -98,6 +98,37 @@ mod exception {
             Self::MSG.to_string()
         }
     }
+
+    #[gen_stub_pyclass]
+    #[pyclass(frozen, extends=PyException, name="MessageTooLargeError")]
+    pub struct PyMessageTooLargeError {}
+
+    impl PyMessageTooLargeError {
+        const MSG: &'static str = "Gossipsub message exceeds max_transmit_size. Reduce prompt length or increase the limit.";
+
+        pub(crate) fn new_err() -> PyErr {
+            PyErr::new::<Self, _>(())
+        }
+    }
+
+    #[gen_stub_pymethods]
+    #[pymethods]
+    impl PyMessageTooLargeError {
+        #[new]
+        #[pyo3(signature = (*args))]
+        #[allow(unused_variables)]
+        pub(crate) fn new(args: &Bound<'_, PyTuple>) -> Self {
+            Self {}
+        }
+
+        fn __repr__(&self) -> String {
+            format!("MessageTooLargeError(\"{}\")", Self::MSG)
+        }
+
+        fn __str__(&self) -> String {
+            Self::MSG.to_string()
+        }
+    }
 }
 
 /// Connection or disconnection event discriminant type.
@@ -200,6 +231,8 @@ async fn networking_task(
                             Err(exception::PyNoPeersSubscribedToTopicError::new_err())
                         } else if let Err(PublishError::AllQueuesFull(_)) = result {
                             Err(exception::PyAllQueuesFullError::new_err())
+                        } else if let Err(PublishError::MessageTooLarge) = result {
+                            Err(exception::PyMessageTooLargeError::new_err())
                         } else {
                             result.pyerr()
                         };
@@ -561,6 +594,7 @@ impl PyNetworkingHandle {
 pub fn networking_submodule(m: &Bound<'_, PyModule>) -> PyResult<()> {
     m.add_class::<exception::PyNoPeersSubscribedToTopicError>()?;
     m.add_class::<exception::PyAllQueuesFullError>()?;
+    m.add_class::<exception::PyMessageTooLargeError>()?;
 
     m.add_class::<PyConnectionUpdateType>()?;
     m.add_class::<PyConnectionUpdate>()?;

src/exo/download/coordinator.py

@@ -5,7 +5,6 @@ from random import random
 
 import anyio
 from anyio import current_time
-from anyio.abc import TaskGroup
 from loguru import logger
 
 from exo.download.download_utils import (
@@ -41,6 +40,7 @@ from exo.shared.types.worker.downloads import (
 )
 from exo.shared.types.worker.shards import PipelineShardMetadata, ShardMetadata
 from exo.utils.channels import Receiver, Sender, channel
+from exo.utils.task_group import TaskGroup
 
 
 @dataclass
@@ -66,7 +66,7 @@ class DownloadCoordinator:
     # Internal event channel for forwarding (initialized in __post_init__)
     event_sender: Sender[Event] = field(init=False)
     event_receiver: Receiver[Event] = field(init=False)
-    _tg: TaskGroup = field(init=False, default_factory=anyio.create_task_group)
+    _tg: TaskGroup = field(init=False, default_factory=TaskGroup)
 
     # Per-model throttle for download progress events
     _last_progress_time: dict[ModelId, float] = field(default_factory=dict)
@@ -167,7 +167,7 @@ class DownloadCoordinator:
                 self._tg.start_soon(self._emit_existing_download_progress)
 
     def shutdown(self) -> None:
-        self._tg.cancel_scope.cancel()
+        self._tg.cancel_tasks()
 
     # directly copied from worker
     async def _resend_out_for_delivery(self) -> None:

src/exo/download/download_utils.py

@@ -823,6 +823,7 @@ async def download_shard(
 
     for file in filtered_file_list:
         downloaded_bytes = await get_downloaded_size(target_dir / file.path)
+        final_file_exists = await aios.path.exists(target_dir / file.path)
         file_progress[file.path] = RepoFileDownloadProgress(
             repo_id=shard.model_card.model_id,
             repo_revision=revision,
@@ -832,7 +833,9 @@ async def download_shard(
             total=Memory.from_bytes(file.size or 0),
             speed=0,
             eta=timedelta(0),
-            status="complete" if downloaded_bytes == file.size else "not_started",
+            status="complete"
+            if final_file_exists and downloaded_bytes == file.size
+            else "not_started",
             start_time=time.time(),
         )
 

src/exo/main.py

@@ -7,7 +7,6 @@ from dataclasses import dataclass, field
 from typing import Self
 
 import anyio
-from anyio.abc import TaskGroup
 from loguru import logger
 from pydantic import PositiveInt
 
@@ -23,6 +22,7 @@ from exo.shared.logging import logger_cleanup, logger_setup
 from exo.shared.types.common import NodeId, SessionId
 from exo.utils.channels import Receiver, channel
 from exo.utils.pydantic_ext import CamelCaseModel
+from exo.utils.task_group import TaskGroup
 from exo.worker.main import Worker
 
 
@@ -38,7 +38,7 @@ class Node:
 
     node_id: NodeId
     offline: bool
-    _tg: TaskGroup = field(init=False, default_factory=anyio.create_task_group)
+    _tg: TaskGroup = field(init=False, default_factory=TaskGroup)
 
     @classmethod
     async def create(cls, args: "Args") -> Self:
@@ -149,11 +149,11 @@ class Node:
 
     def shutdown(self):
         # if this is our second call to shutdown, just sys.exit
-        if self._tg.cancel_scope.cancel_called:
+        if self._tg.cancel_called():
             import sys
 
             sys.exit(1)
-        self._tg.cancel_scope.cancel()
+        self._tg.cancel_tasks()
 
     async def _elect_loop(self):
         with self.election_result_receiver as results:
@@ -252,7 +252,7 @@ def main():
     target = min(max(soft, 65535), hard)
     resource.setrlimit(resource.RLIMIT_NOFILE, (target, hard))
 
-    mp.set_start_method("spawn")
+    mp.set_start_method("spawn", force=True)
     # TODO: Refactor the current verbosity system
     logger_setup(EXO_LOG, args.verbosity)
     logger.info("Starting EXO")

src/exo/master/api.py

@@ -11,8 +11,7 @@ from typing import Annotated, Literal, cast
 from uuid import uuid4
 
 import anyio
-from anyio import BrokenResourceError, create_task_group
-from anyio.abc import TaskGroup
+from anyio import BrokenResourceError
 from fastapi import FastAPI, File, Form, HTTPException, Query, Request, UploadFile
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import FileResponse, JSONResponse, StreamingResponse
@@ -51,6 +50,7 @@ from exo.master.placement import place_instance as get_instance_placements
 from exo.shared.apply import apply
 from exo.shared.constants import (
     DASHBOARD_DIR,
+    EXO_CACHE_HOME,
     EXO_EVENT_LOG_DIR,
     EXO_IMAGE_CACHE_DIR,
     EXO_MAX_CHUNK_SIZE,
@@ -173,8 +173,10 @@ from exo.shared.types.worker.shards import Sharding
 from exo.utils.banner import print_startup_banner
 from exo.utils.channels import Receiver, Sender, channel
 from exo.utils.event_buffer import OrderedBuffer
+from exo.utils.task_group import TaskGroup
 
 _API_EVENT_LOG_DIR = EXO_EVENT_LOG_DIR / "api"
+ONBOARDING_COMPLETE_FILE = EXO_CACHE_HOME / "onboarding_complete"
 
 
 def _format_to_content_type(image_format: Literal["png", "jpeg", "webp"] | None) -> str:
@@ -250,7 +252,7 @@ class API:
             CommandId, Sender[ImageChunk | ErrorChunk]
         ] = {}
         self._image_store = ImageStore(EXO_IMAGE_CACHE_DIR)
-        self._tg: TaskGroup | None = None
+        self._tg: TaskGroup = TaskGroup()
 
     def reset(self, new_session_id: SessionId, result_clock: int):
         logger.info("Resetting API State")
@@ -345,6 +347,8 @@ class API:
         self.app.get("/v1/traces/{task_id}")(self.get_trace)
         self.app.get("/v1/traces/{task_id}/stats")(self.get_trace_stats)
         self.app.get("/v1/traces/{task_id}/raw")(self.get_trace_raw)
+        self.app.get("/onboarding")(self.get_onboarding)
+        self.app.post("/onboarding")(self.complete_onboarding)
 
     async def place_instance(self, payload: PlaceInstanceParams):
         command = PlaceInstance(
@@ -1587,8 +1591,7 @@ class API:
         shutdown_ev = anyio.Event()
 
         try:
-            async with create_task_group() as tg:
-                self._tg = tg
+            async with self._tg as tg:
                 logger.info("Starting API")
                 tg.start_soon(self._apply_state)
                 tg.start_soon(self._pause_on_new_election)
@@ -1814,3 +1817,11 @@ class API:
             media_type="application/json",
             filename=f"trace_{task_id}.json",
         )
+
+    async def get_onboarding(self) -> JSONResponse:
+        return JSONResponse({"completed": ONBOARDING_COMPLETE_FILE.exists()})
+
+    async def complete_onboarding(self) -> JSONResponse:
+        ONBOARDING_COMPLETE_FILE.parent.mkdir(parents=True, exist_ok=True)
+        ONBOARDING_COMPLETE_FILE.write_text("true")
+        return JSONResponse({"completed": True})

src/exo/master/main.py

@@ -1,7 +1,6 @@
 from datetime import datetime, timedelta, timezone
 
 import anyio
-from anyio.abc import TaskGroup
 from loguru import logger
 
 from exo.master.event_log import DiskEventLog
@@ -36,8 +35,6 @@ from exo.shared.types.events import (
     IndexedEvent,
     InputChunkReceived,
     InstanceDeleted,
-    JacclSideChannelData,
-    JacclSideChannelGathered,
     LocalForwarderEvent,
     NodeGatheredInfo,
     NodeTimedOut,
@@ -63,9 +60,9 @@ from exo.shared.types.tasks import (
     TextGeneration as TextGenerationTask,
 )
 from exo.shared.types.worker.instances import InstanceId
-from exo.shared.types.worker.runners import RunnerId
 from exo.utils.channels import Receiver, Sender, channel
 from exo.utils.event_buffer import MultiSourceBuffer
+from exo.utils.task_group import TaskGroup
 
 
 class Master:
@@ -80,7 +77,7 @@ class Master:
         download_command_sender: Sender[ForwarderDownloadCommand],
     ):
         self.state = State()
-        self._tg: TaskGroup = anyio.create_task_group()
+        self._tg: TaskGroup = TaskGroup()
         self.node_id = node_id
         self.session_id = session_id
         self.command_task_mapping: dict[CommandId, TaskId] = {}
@@ -99,7 +96,6 @@ class Master:
         self._event_log = DiskEventLog(EXO_EVENT_LOG_DIR / "master")
         self._pending_traces: dict[TaskId, dict[int, list[TraceEventData]]] = {}
         self._expected_ranks: dict[TaskId, set[int]] = {}
-        self._jaccl_pending: dict[InstanceId, dict[int, dict[RunnerId, bytes]]] = {}
 
     async def run(self):
         logger.info("Starting Master")
@@ -120,7 +116,7 @@ class Master:
 
     async def shutdown(self):
         logger.info("Stopping Master")
-        self._tg.cancel_scope.cancel()
+        self._tg.cancel_tasks()
 
     async def _command_processor(self) -> None:
         with self.command_receiver as commands:
@@ -413,11 +409,6 @@ class Master:
                     self._event_log.append(event)
                     await self._send_event(indexed)
 
-                    # After broadcasting JacclSideChannelData, accumulate and
-                    # emit gathered result when all runners have contributed.
-                    if isinstance(event, JacclSideChannelData):
-                        await self._handle_jaccl_side_channel(event)
-
     async def _loopback_processor(self) -> None:
         # this would ideally not be necessary.
         # this is WAY less hacky than how I was working around this before
@@ -471,42 +462,3 @@ class Master:
         del self._pending_traces[task_id]
         if task_id in self._expected_ranks:
             del self._expected_ranks[task_id]
-
-    async def _handle_jaccl_side_channel(self, event: JacclSideChannelData) -> None:
-        """Accumulate SideChannel contributions; when all runners for an instance
-        have submitted for the same sequence, emit JacclSideChannelGathered."""
-        iid = event.instance_id
-        seq = event.sequence
-
-        if iid not in self._jaccl_pending:
-            self._jaccl_pending[iid] = {}
-        if seq not in self._jaccl_pending[iid]:
-            self._jaccl_pending[iid][seq] = {}
-        self._jaccl_pending[iid][seq][event.runner_id] = event.data
-
-        instance = self.state.instances.get(iid)
-        if instance is None:
-            logger.warning(f"JacclSideChannelData for unknown instance {iid}")
-            return
-
-        expected_runners = set(instance.shard_assignments.runner_to_shard.keys())
-        submitted = set(self._jaccl_pending[iid][seq].keys())
-
-        logger.info(
-            f"JACCL side channel: instance={iid} seq={seq} "
-            f"submitted={len(submitted)}/{len(expected_runners)}"
-        )
-
-        if submitted >= expected_runners:
-            gathered = dict(self._jaccl_pending[iid][seq])
-            del self._jaccl_pending[iid][seq]
-            if not self._jaccl_pending[iid]:
-                del self._jaccl_pending[iid]
-
-            await self.event_sender.send(
-                JacclSideChannelGathered(
-                    instance_id=iid,
-                    sequence=seq,
-                    gathered_data=gathered,
-                )
-            )

src/exo/master/placement.py

@@ -112,7 +112,11 @@ def place_instance(
         cycle for cycle in smallest_cycles if topology.is_rdma_cycle(cycle)
     ]
 
-    if command.instance_meta == InstanceMeta.MlxJaccl and smallest_rdma_cycles != []:
+    if command.instance_meta == InstanceMeta.MlxJaccl:
+        if not smallest_rdma_cycles:
+            raise ValueError(
+                "Requested RDMA (MlxJaccl) but no RDMA-connected cycles available"
+            )
         smallest_cycles = smallest_rdma_cycles
 
     cycles_with_leaf_nodes: list[Cycle] = [
@@ -129,6 +133,11 @@ def place_instance(
         ),
     )
 
+    # Single-node: force Pipeline/Ring (Tensor and Jaccl require multi-node)
+    if len(selected_cycle) == 1:
+        command.instance_meta = InstanceMeta.MlxRing
+        command.sharding = Sharding.Pipeline
+
     shard_assignments = get_shard_assignments(
         command.model_card, selected_cycle, command.sharding, node_memory
     )
@@ -138,9 +147,6 @@ def place_instance(
     instance_id = InstanceId()
     target_instances = dict(deepcopy(current_instances))
 
-    if len(selected_cycle) == 1:
-        command.instance_meta = InstanceMeta.MlxRing
-
     match command.instance_meta:
         case InstanceMeta.MlxJaccl:
             # TODO(evan): shard assignments should contain information about ranks, this is ugly

src/exo/master/tests/test_placement.py

@@ -14,10 +14,12 @@ from exo.shared.models.model_cards import ModelCard, ModelId, ModelTask
 from exo.shared.topology import Topology
 from exo.shared.types.commands import PlaceInstance
 from exo.shared.types.common import CommandId, NodeId
-from exo.shared.types.events import InstanceCreated, InstanceDeleted
+from exo.shared.types.events import InstanceCreated, InstanceDeleted, TaskStatusUpdated
 from exo.shared.types.memory import Memory
 from exo.shared.types.multiaddr import Multiaddr
 from exo.shared.types.profiling import NetworkInterfaceInfo, NodeNetworkInfo
+from exo.shared.types.tasks import TaskId, TaskStatus, TextGeneration
+from exo.shared.types.text_generation import InputMessage, TextGenerationTaskParams
 from exo.shared.types.topology import Connection, SocketConnection
 from exo.shared.types.worker.instances import (
     Instance,
@@ -456,3 +458,117 @@ def test_tensor_rdma_backend_connectivity_matrix(
         else:
             ip_part = coordinator.split(":")[0]
             assert len(ip_part.split(".")) == 4
+
+
+def _make_task(
+    instance_id: InstanceId,
+    status: TaskStatus = TaskStatus.Running,
+) -> TextGeneration:
+    return TextGeneration(
+        task_id=TaskId(),
+        task_status=status,
+        instance_id=instance_id,
+        command_id=CommandId(),
+        task_params=TextGenerationTaskParams(
+            model=ModelId("test-model"),
+            input=[InputMessage(role="user", content="hello")],
+        ),
+    )
+
+
+def test_get_transition_events_delete_instance_cancels_running_tasks(
+    instance: Instance,
+):
+    # arrange
+    instance_id = InstanceId()
+    current_instances: dict[InstanceId, Instance] = {instance_id: instance}
+    target_instances: dict[InstanceId, Instance] = {}
+    task = _make_task(instance_id, TaskStatus.Running)
+    tasks = {task.task_id: task}
+
+    # act
+    events = get_transition_events(current_instances, target_instances, tasks)
+
+    # assert – cancellation event should come before the deletion event
+    assert len(events) == 2
+    assert isinstance(events[0], TaskStatusUpdated)
+    assert events[0].task_id == task.task_id
+    assert events[0].task_status == TaskStatus.Cancelled
+    assert isinstance(events[1], InstanceDeleted)
+    assert events[1].instance_id == instance_id
+
+
+def test_get_transition_events_delete_instance_cancels_pending_tasks(
+    instance: Instance,
+):
+    # arrange
+    instance_id = InstanceId()
+    current_instances: dict[InstanceId, Instance] = {instance_id: instance}
+    target_instances: dict[InstanceId, Instance] = {}
+    task = _make_task(instance_id, TaskStatus.Pending)
+    tasks = {task.task_id: task}
+
+    # act
+    events = get_transition_events(current_instances, target_instances, tasks)
+
+    # assert
+    assert len(events) == 2
+    assert isinstance(events[0], TaskStatusUpdated)
+    assert events[0].task_id == task.task_id
+    assert events[0].task_status == TaskStatus.Cancelled
+    assert isinstance(events[1], InstanceDeleted)
+
+
+def test_get_transition_events_delete_instance_ignores_completed_tasks(
+    instance: Instance,
+):
+    # arrange
+    instance_id = InstanceId()
+    current_instances: dict[InstanceId, Instance] = {instance_id: instance}
+    target_instances: dict[InstanceId, Instance] = {}
+    tasks = {
+        t.task_id: t
+        for t in [
+            _make_task(instance_id, TaskStatus.Complete),
+            _make_task(instance_id, TaskStatus.Failed),
+            _make_task(instance_id, TaskStatus.TimedOut),
+            _make_task(instance_id, TaskStatus.Cancelled),
+        ]
+    }
+
+    # act
+    events = get_transition_events(current_instances, target_instances, tasks)
+
+    # assert – only the InstanceDeleted event, no cancellations
+    assert len(events) == 1
+    assert isinstance(events[0], InstanceDeleted)
+
+
+def test_get_transition_events_delete_instance_cancels_only_matching_tasks(
+    instance: Instance,
+):
+    # arrange
+    instance_id_a = InstanceId()
+    instance_id_b = InstanceId()
+    current_instances: dict[InstanceId, Instance] = {
+        instance_id_a: instance,
+        instance_id_b: instance,
+    }
+    # only delete instance A, keep instance B
+    target_instances: dict[InstanceId, Instance] = {instance_id_b: instance}
+
+    task_a = _make_task(instance_id_a, TaskStatus.Running)
+    task_b = _make_task(instance_id_b, TaskStatus.Running)
+    tasks = {task_a.task_id: task_a, task_b.task_id: task_b}
+
+    # act
+    events = get_transition_events(current_instances, target_instances, tasks)
+
+    # assert – only task_a should be cancelled
+    cancel_events = [e for e in events if isinstance(e, TaskStatusUpdated)]
+    delete_events = [e for e in events if isinstance(e, InstanceDeleted)]
+    assert len(cancel_events) == 1
+    assert cancel_events[0].task_id == task_a.task_id
+    assert cancel_events[0].task_status == TaskStatus.Cancelled
+    assert len(delete_events) == 1
+    assert delete_events[0].instance_id == instance_id_a

src/exo/routing/router.py

@@ -8,14 +8,13 @@ from typing import cast
 from anyio import (
     BrokenResourceError,
     ClosedResourceError,
-    create_task_group,
     move_on_after,
     sleep_forever,
 )
-from anyio.abc import TaskGroup
 from exo_pyo3_bindings import (
     AllQueuesFullError,
     Keypair,
+    MessageTooLargeError,
     NetworkingHandle,
     NoPeersSubscribedToTopicError,
 )
@@ -25,6 +24,7 @@ from loguru import logger
 from exo.shared.constants import EXO_NODE_ID_KEYPAIR
 from exo.utils.channels import Receiver, Sender, channel
 from exo.utils.pydantic_ext import CamelCaseModel
+from exo.utils.task_group import TaskGroup
 
 from .connection_message import ConnectionMessage
 from .topics import CONNECTION_MESSAGES, PublishPolicy, TypedTopic
@@ -111,10 +111,9 @@ class Router:
         self._net: NetworkingHandle = handle
         self._tmp_networking_sender: Sender[tuple[str, bytes]] | None = send
         self._id_count = count()
-        self._tg: TaskGroup | None = None
+        self._tg: TaskGroup = TaskGroup()
 
     async def register_topic[T: CamelCaseModel](self, topic: TypedTopic[T]):
-        assert self._tg is None, "Attempted to register topic after setup time"
         send = self._tmp_networking_sender
         if send:
             self._tmp_networking_sender = None
@@ -148,8 +147,7 @@ class Router:
     async def run(self):
         logger.debug("Starting Router")
         try:
-            async with create_task_group() as tg:
-                self._tg = tg
+            async with self._tg as tg:
                 for topic in self.topic_routers:
                     router = self.topic_routers[topic]
                     tg.start_soon(router.run)
@@ -165,9 +163,7 @@ class Router:
 
     async def shutdown(self):
         logger.debug("Shutting down Router")
-        if not self._tg:
-            return
-        self._tg.cancel_scope.cancel()
+        self._tg.cancel_tasks()
 
     async def _networking_subscribe(self, topic: str):
         await self._net.gossipsub_subscribe(topic)
@@ -211,6 +207,10 @@ class Router:
                     pass
                 except AllQueuesFullError:
                     logger.warning(f"All peer queues full, dropping message on {topic}")
+                except MessageTooLargeError:
+                    logger.warning(
+                        f"Message too large for gossipsub on {topic} ({len(data)} bytes), dropping"
+                    )
 
 
 def get_node_id_keypair(

src/exo/shared/apply.py

@@ -12,8 +12,6 @@ from exo.shared.types.events import (
     InputChunkReceived,
     InstanceCreated,
     InstanceDeleted,
-    JacclSideChannelData,
-    JacclSideChannelGathered,
     NodeDownloadProgress,
     NodeGatheredInfo,
     NodeTimedOut,
@@ -68,8 +66,6 @@ def event_apply(event: Event, state: State) -> State:
             | InputChunkReceived()
             | TracesCollected()
             | TracesMerged()
-            | JacclSideChannelData()
-            | JacclSideChannelGathered()
         ):  # Pass-through events that don't modify state
             return state
         case InstanceCreated():

src/exo/shared/election.py

@@ -4,10 +4,8 @@ import anyio
 from anyio import (
     CancelScope,
     Event,
-    create_task_group,
     get_cancelled_exc_class,
 )
-from anyio.abc import TaskGroup
 from loguru import logger
 
 from exo.routing.connection_message import ConnectionMessage
@@ -15,6 +13,7 @@ from exo.shared.types.commands import ForwarderCommand
 from exo.shared.types.common import NodeId, SessionId
 from exo.utils.channels import Receiver, Sender
 from exo.utils.pydantic_ext import CamelCaseModel
+from exo.utils.task_group import TaskGroup
 
 DEFAULT_ELECTION_TIMEOUT = 3.0
 
@@ -82,13 +81,12 @@ class Election:
         self._candidates: list[ElectionMessage] = []
         self._campaign_cancel_scope: CancelScope | None = None
         self._campaign_done: Event | None = None
-        self._tg: TaskGroup | None = None
+        self._tg = TaskGroup()
 
     async def run(self):
         logger.info("Starting Election")
         try:
-            async with create_task_group() as tg:
-                self._tg = tg
+            async with self._tg as tg:
                 tg.start_soon(self._election_receiver)
                 tg.start_soon(self._connection_receiver)
                 tg.start_soon(self._command_counter)
@@ -124,12 +122,7 @@ class Election:
         )
 
     async def shutdown(self) -> None:
-        if not self._tg:
-            logger.warning(
-                "Attempted to shutdown election service that was not running"
-            )
-            return
-        self._tg.cancel_scope.cancel()
+        self._tg.cancel_tasks()
 
     async def _election_receiver(self) -> None:
         with self._em_receiver as election_messages:
@@ -143,7 +136,6 @@ class Election:
                 if message.clock > self.clock:
                     self.clock = message.clock
                     logger.debug(f"New clock: {self.clock}")
-                    assert self._tg is not None
                     logger.debug("Starting new campaign")
                     candidates: list[ElectionMessage] = [message]
                     logger.debug(f"Candidates: {candidates}")
@@ -178,7 +170,6 @@ class Election:
                 # These messages are strictly peer to peer
                 self.clock += 1
                 logger.debug(f"New clock: {self.clock}")
-                assert self._tg is not None
                 candidates: list[ElectionMessage] = []
                 self._candidates = candidates
                 logger.debug("Starting new campaign")

src/exo/shared/models/model_cards.py

@@ -90,6 +90,7 @@ class ModelCard(CamelCaseModel):
     base_model: str = ""
     capabilities: list[str] = []
     uses_cfg: bool = False
+    trust_remote_code: bool = True
 
     @field_validator("tasks", mode="before")
     @classmethod
@@ -137,6 +138,7 @@ class ModelCard(CamelCaseModel):
             hidden_size=config_data.hidden_size or 0,
             supports_tensor=config_data.supports_tensor,
             tasks=[ModelTask.TextGeneration],
+            trust_remote_code=False,
         )
         await mc.save_to_custom_dir()
         _card_cache[model_id] = mc

src/exo/shared/types/events.py

@@ -1,9 +1,7 @@
-import base64
-from collections.abc import Mapping
 from datetime import datetime
-from typing import Annotated, final
+from typing import final
 
-from pydantic import BeforeValidator, Field, PlainSerializer
+from pydantic import Field
 
 from exo.shared.topology import Connection
 from exo.shared.types.chunks import GenerationChunk, InputImageChunk
@@ -16,28 +14,6 @@ from exo.utils.info_gatherer.info_gatherer import GatheredInfo
 from exo.utils.pydantic_ext import CamelCaseModel, FrozenModel, TaggedModel
 
 
-def _decode_base64_bytes(v: bytes | str) -> bytes:
-    if isinstance(v, bytes):
-        return v
-    return base64.b64decode(v)
-
-
-def _encode_base64_bytes(v: bytes) -> str:
-    return base64.b64encode(v).decode("ascii")
-
-
-Base64Bytes = Annotated[
-    bytes,
-    BeforeValidator(_decode_base64_bytes),
-    PlainSerializer(_encode_base64_bytes, return_type=str),
-]
-"""bytes that serialize to/from base64 strings in JSON.
-
-Needed because TaggedModel's wrap validator converts JSON→Python validation
-context, which breaks strict-mode bytes deserialization from JSON strings.
-"""
-
-
 class EventId(Id):
     """
     Newtype around `ID`
@@ -156,25 +132,6 @@ class TracesMerged(BaseEvent):
     traces: list[TraceEventData]
 
 
-@final
-class JacclSideChannelData(BaseEvent):
-    """A runner's local contribution to a JACCL SideChannel all_gather round."""
-
-    instance_id: InstanceId
-    runner_id: RunnerId
-    sequence: int
-    data: Base64Bytes
-
-
-@final
-class JacclSideChannelGathered(BaseEvent):
-    """Gathered result of a JACCL SideChannel all_gather round."""
-
-    instance_id: InstanceId
-    sequence: int
-    gathered_data: Mapping[RunnerId, Base64Bytes]
-
-
 Event = (
     TestEvent
     | TaskCreated
@@ -195,8 +152,6 @@ Event = (
     | TopologyEdgeDeleted
     | TracesCollected
     | TracesMerged
-    | JacclSideChannelData
-    | JacclSideChannelGathered
 )
 
 

src/exo/shared/types/worker/runners.py

@@ -34,7 +34,8 @@ class RunnerConnected(BaseRunnerStatus):
 
 
 class RunnerLoading(BaseRunnerStatus):
-    pass
+    layers_loaded: int = 0
+    total_layers: int = 0
 
 
 class RunnerLoaded(BaseRunnerStatus):

src/exo/utils/banner.py

@@ -1,8 +1,27 @@
+import logging
+import os
 import sys
+import webbrowser
+
+from exo.shared.constants import EXO_CONFIG_HOME
+
+logger = logging.getLogger(__name__)
+
+_FIRST_RUN_MARKER = EXO_CONFIG_HOME / ".dashboard_opened"
+
+
+def _is_first_run() -> bool:
+    return not _FIRST_RUN_MARKER.exists()
+
+
+def _mark_first_run_done() -> None:
+    _FIRST_RUN_MARKER.parent.mkdir(parents=True, exist_ok=True)
+    _FIRST_RUN_MARKER.touch()
 
 
 def print_startup_banner(port: int) -> None:
     dashboard_url = f"http://localhost:{port}"
+    first_run = _is_first_run()
     banner = f"""
 ╔═══════════════════════════════════════════════════════════════════════╗
 ║                                                                       ║
@@ -30,3 +49,14 @@ def print_startup_banner(port: int) -> None:
 """
 
     print(banner, file=sys.stderr)
+
+    if first_run:
+        # Skip browser open when running inside the native macOS app —
+        # FirstLaunchPopout.swift handles the auto-open with a countdown.
+        if not os.environ.get("EXO_RUNTIME_DIR"):
+            try:
+                webbrowser.open(dashboard_url)
+                logger.info("First run detected — opening dashboard in browser")
+            except Exception:
+                logger.debug("Could not auto-open browser", exc_info=True)
+        _mark_first_run_done()

src/exo/utils/channels.py

@@ -5,7 +5,7 @@ from math import inf
 from multiprocessing.synchronize import Event
 from queue import Empty, Full
 from types import TracebackType
-from typing import Self
+from typing import Any, Self
 
 from anyio import (
     CapacityLimiter,
@@ -157,7 +157,7 @@ class MpSender[T]:
     ) -> None:
         self.close()
 
-    def __getstate__(self):
+    def __getstate__(self) -> dict[str, Any]:
         d = self.__dict__.copy()
         d.pop("__orig_class__", None)
         return d

src/exo/utils/info_gatherer/info_gatherer.py

@@ -8,8 +8,7 @@ from subprocess import CalledProcessError
 from typing import Self, cast
 
 import anyio
-from anyio import create_task_group, fail_after, open_process, to_thread
-from anyio.abc import TaskGroup
+from anyio import fail_after, open_process, to_thread
 from anyio.streams.buffered import BufferedByteReceiveStream
 from anyio.streams.text import TextReceiveStream
 from loguru import logger
@@ -30,6 +29,7 @@ from exo.shared.types.thunderbolt import (
 )
 from exo.utils.channels import Sender
 from exo.utils.pydantic_ext import TaggedModel
+from exo.utils.task_group import TaskGroup
 
 from .macmon import MacmonMetrics
 from .system_info import (
@@ -381,7 +381,7 @@ class InfoGatherer:
     static_info_poll_interval: float | None = 60
     rdma_ctl_poll_interval: float | None = 10 if IS_DARWIN else None
     disk_poll_interval: float | None = 30
-    _tg: TaskGroup = field(init=False, default_factory=create_task_group)
+    _tg: TaskGroup = field(init=False, default_factory=TaskGroup)
 
     async def run(self):
         async with self._tg as tg:
@@ -408,7 +408,7 @@ class InfoGatherer:
                 await self.info_sender.send(nc)
 
     def shutdown(self):
-        self._tg.cancel_scope.cancel()
+        self._tg.cancel_tasks()
 
     async def _monitor_static_info(self):
         if self.static_info_poll_interval is None:

src/exo/utils/task_group.py

@@ -0,0 +1,65 @@
+from collections.abc import Awaitable, Callable
+from dataclasses import dataclass, field
+from types import TracebackType
+from typing import Any, Unpack
+
+from anyio import create_task_group
+from anyio.abc import TaskGroup as TaskGroupABC
+
+
+@dataclass
+class TaskGroup:
+    _tg: TaskGroupABC | None = field(default=None, init=False)
+    _queued: list[tuple[Any, Any, Any]] | None = field(default_factory=list, init=False)
+
+    def is_running(self) -> bool:
+        return self._tg is not None
+
+    def cancel_tasks(self):
+        assert self._tg
+        self._tg.cancel_scope.cancel()
+
+    def cancel_called(self) -> bool:
+        assert self._tg
+        return self._tg.cancel_scope.cancel_called
+
+    def start_soon[*T](
+        self,
+        func: Callable[[Unpack[T]], Awaitable[Any]],
+        *args: Unpack[T],
+        name: object = None,
+    ) -> None:
+        assert self._tg is not None
+        assert self._queued is None
+        self._tg.start_soon(func, *args, name=name)
+
+    def queue[*T](
+        self,
+        func: Callable[[Unpack[T]], Awaitable[Any]],
+        *args: Unpack[T],
+        name: object = None,
+    ) -> None:
+        assert self._tg is None
+        assert self._queued is not None
+        self._queued.append((func, args, name))
+
+    async def __aenter__(self) -> TaskGroupABC:
+        assert self._tg is None
+        assert self._queued is not None
+        self._tg = create_task_group()
+        r = await self._tg.__aenter__()
+        for func, args, name in self._queued:  # pyright: ignore[reportAny]
+            self._tg.start_soon(func, *args, name=name)  # pyright: ignore[reportAny]
+        self._queued = None
+        return r
+
+    async def __aexit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_val: BaseException | None,
+        exc_tb: TracebackType | None,
+    ) -> bool:
+        """Exit the task group context waiting for all tasks to finish."""
+        assert self._tg is not None, "aenter sets self.lazy, so it exists when we aexit"
+        assert self._queued is None
+        return await self._tg.__aexit__(exc_type, exc_val, exc_tb)

src/exo/worker/engines/mlx/auto_parallel.py

@@ -32,13 +32,12 @@ from mlx_lm.models.minimax import MiniMaxAttention
 from mlx_lm.models.minimax import Model as MiniMaxModel
 from mlx_lm.models.ministral3 import Model as Ministral3Model
 from mlx_lm.models.qwen3_moe import Model as Qwen3MoeModel
-from mlx_lm.models.qwen3_moe import Qwen3MoeSparseMoeBlock
+from mlx_lm.models.qwen3_moe import Qwen3MoeDecoderLayer, Qwen3MoeSparseMoeBlock
 from mlx_lm.models.qwen3_next import Model as Qwen3NextModel
 from mlx_lm.models.qwen3_next import Qwen3NextDecoderLayer, Qwen3NextSparseMoeBlock
 from mlx_lm.models.step3p5 import Model as Step35Model
 from mlx_lm.models.step3p5 import Step3p5MLP as Step35MLP
 from mlx_lm.models.step3p5 import Step3p5Model as Step35InnerModel
-from transformers.models.qwen3.modeling_qwen3 import Qwen3DecoderLayer
 
 from exo.shared.logging import logger
 from exo.shared.types.worker.shards import PipelineShardMetadata
@@ -47,6 +46,7 @@ if TYPE_CHECKING:
     from mlx_lm.models.cache import Cache
 
 TimeoutCallback = Callable[[], None]
+LayerLoadedCallback = Callable[[int, int], None]  # (layers_loaded, total_layers)
 
 
 def eval_with_timeout(
@@ -186,7 +186,7 @@ def set_pipeline_prefill(model: nn.Module, is_prefill: bool) -> None:
             layer.is_prefill = is_prefill
 
 
-def _inner_model(model: nn.Module) -> nn.Module:
+def get_inner_model(model: nn.Module) -> nn.Module:
     inner = getattr(model, "model", None)
     if isinstance(inner, nn.Module):
         return inner
@@ -204,7 +204,7 @@ def _inner_model(model: nn.Module) -> nn.Module:
     raise ValueError("Model must either have a 'model' or 'transformer' attribute")
 
 
-def _get_layers(inner_model_instance: nn.Module) -> list[_LayerCallable]:
+def get_layers(inner_model_instance: nn.Module) -> list[_LayerCallable]:
     # Handle both model.layers and model.h cases
     layers: list[_LayerCallable]
     if hasattr(inner_model_instance, "layers"):
@@ -221,6 +221,7 @@ def pipeline_auto_parallel(
     model: nn.Module,
     group: mx.distributed.Group,
     model_shard_meta: PipelineShardMetadata,
+    on_layer_loaded: LayerLoadedCallback | None,
 ) -> nn.Module:
     """
     Automatically parallelize a model across multiple devices.
@@ -230,16 +231,19 @@ def pipeline_auto_parallel(
     Returns:
     The parallelized model
     """
-    inner_model_instance: nn.Module = _inner_model(model)
+    inner_model_instance: nn.Module = get_inner_model(model)
 
-    layers = _get_layers(inner_model_instance)
+    layers = get_layers(inner_model_instance)
 
     start_layer, end_layer = model_shard_meta.start_layer, model_shard_meta.end_layer
     device_rank, world_size = model_shard_meta.device_rank, model_shard_meta.world_size
 
     layers = layers[start_layer:end_layer]
-    for layer in layers:
+    total = len(layers)
+    for i, layer in enumerate(layers):
         mx.eval(layer)  # type: ignore
+        if on_layer_loaded is not None:
+            on_layer_loaded(i, total)
 
     layers[0] = PipelineFirstLayer(layers[0], device_rank, group=group)
     layers[-1] = PipelineLastLayer(
@@ -351,8 +355,9 @@ def patch_tensor_model[T](model: T) -> T:
 def tensor_auto_parallel(
     model: nn.Module,
     group: mx.distributed.Group,
-    timeout_seconds: float = 60.0,
-    on_timeout: TimeoutCallback | None = None,
+    timeout_seconds: float,
+    on_timeout: TimeoutCallback | None,
+    on_layer_loaded: LayerLoadedCallback | None,
 ) -> nn.Module:
     all_to_sharded_linear = partial(
         shard_linear,
@@ -462,7 +467,7 @@ def tensor_auto_parallel(
         raise ValueError(f"Unsupported model type: {type(model)}")
 
     model = tensor_parallel_sharding_strategy.shard_model(
-        model, timeout_seconds, on_timeout
+        model, timeout_seconds, on_timeout, on_layer_loaded
     )
     return patch_tensor_model(model)
 
@@ -489,6 +494,7 @@ class TensorParallelShardingStrategy(ABC):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module: ...
 
 
@@ -498,13 +504,13 @@ class LlamaShardingStrategy(TensorParallelShardingStrategy):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module:
         model = cast(LlamaModel, model)
-        for layer in model.layers:
+        total = len(model.layers)
+        for i, layer in enumerate(model.layers):
             # Force load weights before sharding to avoid FAST_SYNCH deadlock
-            eval_with_timeout(
-                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
-            )
+            eval_with_timeout(layer.parameters(), timeout_seconds / total, on_timeout)
             layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
             layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
             layer.self_attn.v_proj = self.all_to_sharded_linear(layer.self_attn.v_proj)
@@ -517,11 +523,13 @@ class LlamaShardingStrategy(TensorParallelShardingStrategy):
             layer.mlp.down_proj = self.sharded_to_all_linear(layer.mlp.down_proj)
             layer.mlp.up_proj = self.all_to_sharded_linear(layer.mlp.up_proj)
             mx.eval(layer)
+            if on_layer_loaded is not None:
+                on_layer_loaded(i, total)
         return model
 
 
 def _set_layers(model: nn.Module, layers: list[_LayerCallable]) -> None:
-    inner_model_instance = _inner_model(model)
+    inner_model_instance = get_inner_model(model)
     if hasattr(inner_model_instance, "layers"):
         inner_model_instance.layers = layers
 
@@ -552,13 +560,13 @@ class DeepSeekShardingStrategy(TensorParallelShardingStrategy):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module:
         model = cast(DeepseekV3Model, model)
+        total = len(model.layers)
 
-        for layer in model.layers:
-            eval_with_timeout(
-                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
-            )
+        for i, layer in enumerate(model.layers):
+            eval_with_timeout(layer.parameters(), timeout_seconds / total, on_timeout)
 
             # Shard the self attention
             if layer.self_attn.q_lora_rank is None:
@@ -609,6 +617,8 @@ class DeepSeekShardingStrategy(TensorParallelShardingStrategy):
                 layer.mlp.sharding_group = self.group
 
             mx.eval(layer)
+            if on_layer_loaded is not None:
+                on_layer_loaded(i, total)
 
         return model
 
@@ -635,13 +645,15 @@ class GLM4MoeLiteShardingStrategy(TensorParallelShardingStrategy):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module:
         model = cast(GLM4MoeLiteModel, model)
-        for layer in model.layers:  # type: ignore
+        total = len(model.layers)  # type: ignore
+        for i, layer in enumerate(model.layers):  # type: ignore
             layer = cast(Glm4MoeLiteDecoderLayer, layer)
             eval_with_timeout(
                 layer.parameters(),
-                timeout_seconds / len(model.layers),  # type: ignore
+                timeout_seconds / total,
                 on_timeout,
             )
             if layer.self_attn.q_lora_rank is None:  # type: ignore
@@ -689,6 +701,8 @@ class GLM4MoeLiteShardingStrategy(TensorParallelShardingStrategy):
                 layer.mlp = ShardedMoE(layer.mlp)  # type: ignore
                 layer.mlp.sharding_group = self.group  # type: ignore
             mx.eval(layer)
+            if on_layer_loaded is not None:
+                on_layer_loaded(i, total)
 
         return model
 
@@ -777,12 +791,12 @@ class MiniMaxShardingStrategy(TensorParallelShardingStrategy):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module:
         model = cast(MiniMaxModel, model)
-        for layer in model.layers:
-            eval_with_timeout(
-                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
-            )
+        total = len(model.layers)
+        for i, layer in enumerate(model.layers):
+            eval_with_timeout(layer.parameters(), timeout_seconds / total, on_timeout)
             # Shard the self attention
             layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
             layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
@@ -807,6 +821,8 @@ class MiniMaxShardingStrategy(TensorParallelShardingStrategy):
             layer.block_sparse_moe = ShardedMoE(layer.block_sparse_moe)  # pyright: ignore[reportAttributeAccessIssue, reportArgumentType]
             layer.block_sparse_moe.sharding_group = self.group  # pyright: ignore[reportAttributeAccessIssue]
             mx.eval(layer)
+            if on_layer_loaded is not None:
+                on_layer_loaded(i, total)
         return model
 
 
@@ -816,14 +832,14 @@ class QwenShardingStrategy(TensorParallelShardingStrategy):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module:
         model = cast(Qwen3MoeModel | Qwen3NextModel, model)
-        for layer in model.layers:
-            eval_with_timeout(
-                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
-            )
+        total = len(model.layers)
+        for i, layer in enumerate(model.layers):
+            eval_with_timeout(layer.parameters(), timeout_seconds / total, on_timeout)
             # Shard the self attention
-            if isinstance(layer, Qwen3DecoderLayer):
+            if isinstance(layer, Qwen3MoeDecoderLayer):
                 layer.self_attn.q_proj = self.all_to_sharded_linear(
                     layer.self_attn.q_proj
                 )
@@ -836,6 +852,8 @@ class QwenShardingStrategy(TensorParallelShardingStrategy):
                 layer.self_attn.o_proj = self.sharded_to_all_linear(
                     layer.self_attn.o_proj
                 )
+                layer.self_attn.n_heads //= self.N
+                layer.self_attn.n_kv_heads //= self.N
             else:
                 assert isinstance(layer, Qwen3NextDecoderLayer)
                 if hasattr(layer, "linear_attn"):
@@ -930,6 +948,8 @@ class QwenShardingStrategy(TensorParallelShardingStrategy):
                 layer.mlp.up_proj = self.all_to_sharded_linear(layer.mlp.up_proj)
 
             mx.eval(layer)
+            if on_layer_loaded is not None:
+                on_layer_loaded(i, total)
         return model
 
 
@@ -939,12 +959,12 @@ class Glm4MoeShardingStrategy(TensorParallelShardingStrategy):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module:
         model = cast(Glm4MoeModel, model)
-        for layer in model.layers:
-            eval_with_timeout(
-                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
-            )
+        total = len(model.layers)
+        for i, layer in enumerate(model.layers):
+            eval_with_timeout(layer.parameters(), timeout_seconds / total, on_timeout)
 
             layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
             layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
@@ -976,6 +996,8 @@ class Glm4MoeShardingStrategy(TensorParallelShardingStrategy):
                 layer.mlp.up_proj = self.all_to_sharded_linear(layer.mlp.up_proj)
 
             mx.eval(layer)
+            if on_layer_loaded is not None:
+                on_layer_loaded(i, total)
         return model
 
 
@@ -985,13 +1007,13 @@ class GptOssShardingStrategy(TensorParallelShardingStrategy):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module:
         model = cast(GptOssMoeModel, model)
+        total = len(model.layers)
 
-        for layer in model.layers:
-            eval_with_timeout(
-                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
-            )
+        for i, layer in enumerate(model.layers):
+            eval_with_timeout(layer.parameters(), timeout_seconds / total, on_timeout)
             layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
             layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
             layer.self_attn.v_proj = self.all_to_sharded_linear(layer.self_attn.v_proj)
@@ -1017,6 +1039,8 @@ class GptOssShardingStrategy(TensorParallelShardingStrategy):
             layer.mlp = ShardedMoE(layer.mlp)  # type: ignore
             layer.mlp.sharding_group = self.group  # pyright: ignore[reportAttributeAccessIssue]
             mx.eval(layer)
+            if on_layer_loaded is not None:
+                on_layer_loaded(i, total)
         return model
 
 
@@ -1026,13 +1050,13 @@ class Step35ShardingStrategy(TensorParallelShardingStrategy):
         model: nn.Module,
         timeout_seconds: float,
         on_timeout: TimeoutCallback | None,
+        on_layer_loaded: LayerLoadedCallback | None,
     ) -> nn.Module:
         model = cast(Step35Model, model)
+        total = len(model.layers)
 
-        for layer in model.layers:
-            eval_with_timeout(
-                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
-            )
+        for i, layer in enumerate(model.layers):
+            eval_with_timeout(layer.parameters(), timeout_seconds / total, on_timeout)
             layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
             layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
             layer.self_attn.v_proj = self.all_to_sharded_linear(layer.self_attn.v_proj)
@@ -1060,4 +1084,6 @@ class Step35ShardingStrategy(TensorParallelShardingStrategy):
                 self.sharded_to_all_linear_in_place(layer.mlp.switch_mlp.down_proj)
 
             mx.eval(layer)
+            if on_layer_loaded is not None:
+                on_layer_loaded(i, total)
         return model

src/exo/worker/engines/mlx/utils_mlx.py

@@ -23,9 +23,7 @@ from mlx_lm.models.deepseek_v3 import DeepseekV3Model
 from mlx_lm.tokenizer_utils import TokenizerWrapper
 
 from exo.shared.models.model_cards import ModelId
-from exo.worker.engines.mlx.constants import (
-    TRUST_REMOTE_CODE,
-)
+from exo.worker.engines.mlx.constants import TRUST_REMOTE_CODE
 
 try:
     from mlx_lm.tokenizer_utils import load_tokenizer
@@ -55,8 +53,11 @@ from exo.shared.types.worker.shards import (
 )
 from exo.worker.engines.mlx import Model
 from exo.worker.engines.mlx.auto_parallel import (
+    LayerLoadedCallback,
     TimeoutCallback,
     eval_with_timeout,
+    get_inner_model,
+    get_layers,
     pipeline_auto_parallel,
     tensor_auto_parallel,
 )
@@ -171,13 +172,28 @@ def initialize_mlx(
 def load_mlx_items(
     bound_instance: BoundInstance,
     group: Group | None,
-    on_timeout: TimeoutCallback | None = None,
+    on_timeout: TimeoutCallback | None,
+    on_layer_loaded: LayerLoadedCallback | None,
 ) -> tuple[Model, TokenizerWrapper]:
     if group is None:
         logger.info(f"Single device used for {bound_instance.instance}")
         model_path = build_model_path(bound_instance.bound_shard.model_card.model_id)
         start_time = time.perf_counter()
-        model, _ = load_model(model_path, strict=True)
+        model, _ = load_model(model_path, lazy=True, strict=False)
+        # Eval layers one by one for progress reporting
+        try:
+            inner = get_inner_model(model)
+            layers = get_layers(inner)
+            total = len(layers)
+            for i, layer in enumerate(layers):
+                mx.eval(layer)  # type: ignore
+                if on_layer_loaded is not None:
+                    on_layer_loaded(i, total)
+        except ValueError as e:
+            logger.opt(exception=e).debug(
+                "Model architecture doesn't support layer-by-layer progress tracking",
+            )
+        mx.eval(model)
         end_time = time.perf_counter()
         logger.info(f"Time taken to load model: {(end_time - start_time):.2f}s")
         tokenizer = get_tokenizer(model_path, bound_instance.bound_shard)
@@ -186,7 +202,10 @@ def load_mlx_items(
         logger.info("Starting distributed init")
         start_time = time.perf_counter()
         model, tokenizer = shard_and_load(
-            bound_instance.bound_shard, group=group, on_timeout=on_timeout
+            bound_instance.bound_shard,
+            group=group,
+            on_timeout=on_timeout,
+            on_layer_loaded=on_layer_loaded,
         )
         end_time = time.perf_counter()
         logger.info(
@@ -201,7 +220,8 @@ def load_mlx_items(
 def shard_and_load(
     shard_metadata: ShardMetadata,
     group: Group,
-    on_timeout: TimeoutCallback | None = None,
+    on_timeout: TimeoutCallback | None,
+    on_layer_loaded: LayerLoadedCallback | None,
 ) -> tuple[nn.Module, TokenizerWrapper]:
     model_path = build_model_path(shard_metadata.model_card.model_id)
 
@@ -242,10 +262,14 @@ def shard_and_load(
     match shard_metadata:
         case TensorShardMetadata():
             logger.info(f"loading model from {model_path} with tensor parallelism")
-            model = tensor_auto_parallel(model, group, timeout_seconds, on_timeout)
+            model = tensor_auto_parallel(
+                model, group, timeout_seconds, on_timeout, on_layer_loaded
+            )
         case PipelineShardMetadata():
             logger.info(f"loading model from {model_path} with pipeline parallelism")
-            model = pipeline_auto_parallel(model, group, shard_metadata)
+            model = pipeline_auto_parallel(
+                model, group, shard_metadata, on_layer_loaded=on_layer_loaded
+            )
             eval_with_timeout(model.parameters(), timeout_seconds, on_timeout)
         case CfgShardMetadata():
             raise ValueError(
@@ -267,7 +291,11 @@ def shard_and_load(
 
 def get_tokenizer(model_path: Path, shard_metadata: ShardMetadata) -> TokenizerWrapper:
     """Load tokenizer for a model shard. Delegates to load_tokenizer_for_model_id."""
-    return load_tokenizer_for_model_id(shard_metadata.model_card.model_id, model_path)
+    return load_tokenizer_for_model_id(
+        shard_metadata.model_card.model_id,
+        model_path,
+        trust_remote_code=shard_metadata.model_card.trust_remote_code,
+    )
 
 
 def get_eos_token_ids_for_model(model_id: ModelId) -> list[int] | None:
@@ -299,7 +327,7 @@ def get_eos_token_ids_for_model(model_id: ModelId) -> list[int] | None:
 
 
 def load_tokenizer_for_model_id(
-    model_id: ModelId, model_path: Path
+    model_id: ModelId, model_path: Path, *, trust_remote_code: bool = TRUST_REMOTE_CODE
 ) -> TokenizerWrapper:
     """
     Load tokenizer for a model given its ID and local path.
@@ -368,7 +396,7 @@ def load_tokenizer_for_model_id(
 
     tokenizer = load_tokenizer(
         model_path,
-        tokenizer_config_extra={"trust_remote_code": TRUST_REMOTE_CODE},
+        tokenizer_config_extra={"trust_remote_code": trust_remote_code},
         eos_token_ids=eos_token_ids,
     )
 
@@ -616,7 +644,7 @@ class NullKVCache(KVCache):
         raise NotImplementedError("We should not be setting a NullKVCache.")
 
 
-def mlx_force_oom(size: int = 40000) -> None:
+def mlx_force_oom(size: int = 200000) -> None:
     """
     Force an Out-Of-Memory (OOM) error in MLX by performing large tensor operations.
     """
@@ -643,7 +671,7 @@ def set_wired_limit_for_model(model_size: Memory):
         return
 
     max_rec_size = Memory.from_bytes(
-        int(mx.metal.device_info()["max_recommended_working_set_size"])
+        int(mx.device_info()["max_recommended_working_set_size"])
     )
     if model_size > 0.9 * max_rec_size:
         logger.warning(
@@ -667,11 +695,6 @@ def mlx_cleanup(
 
 
 def mx_any(bool_: bool, group: Group | None) -> bool:
-    """Synchronize a boolean across all distributed nodes.
-
-    Returns True if any node has bool_=True.  Uses all_sum so every
-    node participates in the collective — preventing GPU deadlocks.
-    """
     if group is None:
         return bool_
     num_true = mx.distributed.all_sum(

src/exo/worker/main.py

@@ -3,8 +3,7 @@ from datetime import datetime, timezone
 from random import random
 
 import anyio
-from anyio import CancelScope, create_task_group, fail_after
-from anyio.abc import TaskGroup
+from anyio import CancelScope, fail_after
 from loguru import logger
 
 from exo.download.download_utils import resolve_model_in_path
@@ -24,7 +23,6 @@ from exo.shared.types.events import (
     GlobalForwarderEvent,
     IndexedEvent,
     InputChunkReceived,
-    JacclSideChannelGathered,
     LocalForwarderEvent,
     NodeDownloadProgress,
     NodeGatheredInfo,
@@ -52,6 +50,7 @@ from exo.utils.event_buffer import OrderedBuffer
 from exo.utils.info_gatherer.info_gatherer import GatheredInfo, InfoGatherer
 from exo.utils.info_gatherer.net_profile import check_reachable
 from exo.utils.keyed_backoff import KeyedBackoff
+from exo.utils.task_group import TaskGroup
 from exo.worker.plan import plan
 from exo.worker.runner.runner_supervisor import RunnerSupervisor
 
@@ -81,7 +80,7 @@ class Worker:
 
         self.state: State = State()
         self.runners: dict[RunnerId, RunnerSupervisor] = {}
-        self._tg: TaskGroup = create_task_group()
+        self._tg: TaskGroup = TaskGroup()
 
         self._nack_cancel_scope: CancelScope | None = None
         self._nack_attempts: int = 0
@@ -165,15 +164,6 @@ class Worker:
                 for idx, event in indexed_events:
                     self.state = apply(self.state, IndexedEvent(idx=idx, event=event))
 
-                    # Dispatch JACCL gathered events to the relevant RunnerSupervisor
-                    if isinstance(event, JacclSideChannelGathered):
-                        for runner in self.runners.values():
-                            if (
-                                runner.bound_instance.instance.instance_id
-                                == event.instance_id
-                            ):
-                                runner.notify_gathered(event)
-
                     # Buffer input image chunks for image editing
                     if isinstance(event, InputChunkReceived):
                         cmd_id = event.command_id
@@ -327,7 +317,7 @@ class Worker:
                     await self._start_runner_task(task)
 
     def shutdown(self):
-        self._tg.cancel_scope.cancel()
+        self._tg.cancel_tasks()
 
     async def _start_runner_task(self, task: Task):
         if (instance := self.state.instances.get(task.instance_id)) is not None:

src/exo/worker/runner/bootstrap.py

@@ -4,7 +4,7 @@ import loguru
 
 from exo.shared.types.events import Event, RunnerStatusUpdated
 from exo.shared.types.tasks import Task, TaskId
-from exo.shared.types.worker.instances import BoundInstance, MlxJacclInstance
+from exo.shared.types.worker.instances import BoundInstance
 from exo.shared.types.worker.runners import RunnerFailed
 from exo.utils.channels import ClosedResourceError, MpReceiver, MpSender
 
@@ -17,33 +17,16 @@ def entrypoint(
     task_receiver: MpReceiver[Task],
     cancel_receiver: MpReceiver[TaskId],
     _logger: "loguru.Logger",
-    pipe_fifo_paths: tuple[str, str] | None = None,
 ) -> None:
+    global logger
+    logger = _logger
+
     fast_synch_override = os.environ.get("EXO_FAST_SYNCH")
-    if fast_synch_override == "on" or (
-        fast_synch_override != "off"
-        and (
-            isinstance(bound_instance.instance, MlxJacclInstance)
-            and len(bound_instance.instance.jaccl_devices) >= 2
-        )
-    ):
+    if fast_synch_override != "off":
         os.environ["MLX_METAL_FAST_SYNCH"] = "1"
     else:
         os.environ["MLX_METAL_FAST_SYNCH"] = "0"
 
-    # Open JACCL FIFOs by path and set env vars for C++ SideChannel.
-    # Named pipes (FIFOs) work across multiprocessing spawn (macOS default).
-    if pipe_fifo_paths is not None:
-        fifo_c2p, fifo_p2c = pipe_fifo_paths
-        # C++ reads gathered data from p2c (PIPE_IN), writes local data to c2p (PIPE_OUT)
-        pipe_in_fd = os.open(fifo_p2c, os.O_RDONLY)
-        pipe_out_fd = os.open(fifo_c2p, os.O_WRONLY)
-        os.environ["MLX_JACCL_PIPE_IN"] = str(pipe_in_fd)
-        os.environ["MLX_JACCL_PIPE_OUT"] = str(pipe_out_fd)
-
-    global logger
-    logger = _logger
-
     logger.info(f"Fast synch flag: {os.environ['MLX_METAL_FAST_SYNCH']}")
 
     # Import main after setting global logger - this lets us just import logger from this module

src/exo/worker/runner/llm_inference/runner.py

@@ -150,7 +150,10 @@ def main(
                 case LoadModel() if (
                     isinstance(current_status, RunnerConnected) and group is not None
                 ) or (isinstance(current_status, RunnerIdle) and group is None):
-                    current_status = RunnerLoading()
+                    total_layers = shard_metadata.end_layer - shard_metadata.start_layer
+                    current_status = RunnerLoading(
+                        layers_loaded=0, total_layers=total_layers
+                    )
                     logger.info("runner loading")
                     event_sender.send(
                         RunnerStatusUpdated(
@@ -170,11 +173,26 @@ def main(
                         )
                         time.sleep(0.5)
 
+                    def on_layer_loaded(layers_loaded: int, total: int) -> None:
+                        nonlocal current_status
+                        current_status = RunnerLoading(
+                            layers_loaded=layers_loaded, total_layers=total
+                        )
+                        event_sender.send(
+                            RunnerStatusUpdated(
+                                runner_id=runner_id,
+                                runner_status=current_status,
+                            )
+                        )
+
                     assert (
                         ModelTask.TextGeneration in shard_metadata.model_card.tasks
                     ), f"Incorrect model task(s): {shard_metadata.model_card.tasks}"
                     inference_model, tokenizer = load_mlx_items(
-                        bound_instance, group, on_timeout=on_model_load_timeout
+                        bound_instance,
+                        group,
+                        on_timeout=on_model_load_timeout,
+                        on_layer_loaded=on_layer_loaded,
                     )
                     logger.info(
                         f"model has_tool_calling={tokenizer.has_tool_calling} using tokens {tokenizer.tool_call_start}, {tokenizer.tool_call_end}"

src/exo/worker/runner/llm_inference/tool_parsers.py

@@ -51,11 +51,33 @@ def _parse_json_calls(text: str) -> list[ToolCallItem] | None:
         return None
 
 
+def _try_parse_json(v: str) -> str | dict[str, Any] | list[Any]:
+    stripped = v.strip()
+    if (stripped.startswith("[") and stripped.endswith("]")) or (
+        stripped.startswith("{") and stripped.endswith("}")
+    ):
+        try:
+            parsed: dict[str, Any] | list[Any] = json.loads(stripped)  # pyright: ignore[reportAny]
+            return parsed
+        except (json.JSONDecodeError, ValueError):
+            pass
+    return v
+
+
 def _flatten(p: dict[str, Any]) -> dict[str, str]:
-    return {
-        k: json.dumps(v) if isinstance(v, (dict, list)) else str(v)  # pyright: ignore[reportAny]
-        for k, v in p.items()  # pyright: ignore[reportAny]
-    }
+    result: dict[str, str] = {}
+    for k, v in p.items():  # pyright: ignore[reportAny]
+        if isinstance(v, dict):
+            resolved: dict[str, Any] = {
+                str(ik): _try_parse_json(str(iv)) if isinstance(iv, str) else iv  # pyright: ignore[reportUnknownArgumentType]
+                for ik, iv in v.items()  # pyright: ignore[reportUnknownVariableType]
+            }
+            result[k] = json.dumps(resolved)
+        elif isinstance(v, list):
+            result[k] = json.dumps(v)
+        else:
+            result[k] = str(v)  # pyright: ignore[reportAny]
+    return result
 
 
 json_tool_parser = ToolParser(

src/exo/worker/runner/runner_supervisor.py

@@ -1,11 +1,7 @@
 import contextlib
-import os
+import multiprocessing as mp
 import signal
-import struct
-import tempfile
 from dataclasses import dataclass, field
-from functools import partial
-from multiprocessing import Process
 from typing import Self
 
 import anyio
@@ -18,14 +14,12 @@ from loguru import logger
 
 from exo.shared.types.events import (
     Event,
-    JacclSideChannelData,
-    JacclSideChannelGathered,
     RunnerStatusUpdated,
     TaskAcknowledged,
     TaskStatusUpdated,
 )
 from exo.shared.types.tasks import Task, TaskId, TaskStatus
-from exo.shared.types.worker.instances import BoundInstance, MlxJacclInstance
+from exo.shared.types.worker.instances import BoundInstance
 from exo.shared.types.worker.runners import (
     RunnerConnecting,
     RunnerFailed,
@@ -38,28 +32,9 @@ from exo.shared.types.worker.runners import (
 )
 from exo.shared.types.worker.shards import ShardMetadata
 from exo.utils.channels import MpReceiver, MpSender, Sender, mp_channel
+from exo.utils.task_group import TaskGroup
 from exo.worker.runner.bootstrap import entrypoint
 
-
-def _pipe_read_exact(fd: int, n: int) -> bytes | None:
-    """Read exactly n bytes from a file descriptor. Returns None on EOF."""
-    data = b""
-    while len(data) < n:
-        chunk = os.read(fd, n - len(data))
-        if not chunk:
-            return None
-        data += chunk
-    return data
-
-
-def _pipe_write_all(fd: int, data: bytes) -> None:
-    """Write all bytes to a file descriptor."""
-    view = memoryview(data)
-    while view:
-        written = os.write(fd, view)
-        view = view[written:]
-
-
 PREFILL_TIMEOUT_SECONDS = 60
 DECODE_TIMEOUT_SECONDS = 5
 
@@ -68,25 +43,20 @@ DECODE_TIMEOUT_SECONDS = 5
 class RunnerSupervisor:
     shard_metadata: ShardMetadata
     bound_instance: BoundInstance
-    runner_process: Process
+    runner_process: mp.Process
     initialize_timeout: float
     _ev_recv: MpReceiver[Event]
     _task_sender: MpSender[Task]
     _event_sender: Sender[Event]
     _cancel_sender: MpSender[TaskId]
-    _pipe_read_fd: int | None = None  # Python reads runner's pipe output
-    _pipe_write_fd: int | None = None  # Python writes gathered data to runner
-    _child_pipe_fds: tuple[int, int] | None = None  # fds to close after fork
-    _fifo_dir: str | None = None  # Temp dir for FIFO files (for cleanup)
-    _fifo_c2p: str | None = None  # FIFO path: C++ writes → Python reads
-    _fifo_p2c: str | None = None  # FIFO path: Python writes → C++ reads
+    _tg: TaskGroup = field(default_factory=TaskGroup, init=False)
     status: RunnerStatus = field(default_factory=RunnerIdle, init=False)
     pending: dict[TaskId, anyio.Event] = field(default_factory=dict, init=False)
     completed: set[TaskId] = field(default_factory=set, init=False)
     cancelled: set[TaskId] = field(default_factory=set, init=False)
-    _gathered_waiters: dict[
-        int, tuple[anyio.Event, JacclSideChannelGathered | None]
-    ] = field(default_factory=dict, init=False)
+    _cancel_watch_runner: anyio.CancelScope = field(
+        default_factory=anyio.CancelScope, init=False
+    )
 
     @classmethod
     def create(
@@ -100,24 +70,7 @@ class RunnerSupervisor:
         task_sender, task_recv = mp_channel[Task]()
         cancel_sender, cancel_recv = mp_channel[TaskId]()
 
-        # For MlxJaccl instances, create named pipes (FIFOs) for SideChannel relay.
-        # Named pipes work across multiprocessing.Process spawn (macOS default).
-        # FIFO c2p: C++ writes local data → Python reads it
-        # FIFO p2c: Python writes gathered data → C++ reads it
-        fifo_dir: str | None = None
-        fifo_c2p: str | None = None
-        fifo_p2c: str | None = None
-        pipe_fifo_paths: tuple[str, str] | None = None
-
-        if isinstance(bound_instance.instance, MlxJacclInstance):
-            fifo_dir = tempfile.mkdtemp(prefix="exo_jaccl_")
-            fifo_c2p = os.path.join(fifo_dir, "c2p")  # C++ → Python
-            fifo_p2c = os.path.join(fifo_dir, "p2c")  # Python → C++
-            os.mkfifo(fifo_c2p)
-            os.mkfifo(fifo_p2c)
-            pipe_fifo_paths = (fifo_c2p, fifo_p2c)
-
-        runner_process = Process(
+        runner_process = mp.Process(
             target=entrypoint,
             args=(
                 bound_instance,
@@ -125,7 +78,6 @@ class RunnerSupervisor:
                 task_recv,
                 cancel_recv,
                 logger,
-                pipe_fifo_paths,
             ),
             daemon=True,
         )
@@ -141,55 +93,31 @@ class RunnerSupervisor:
             _task_sender=task_sender,
             _cancel_sender=cancel_sender,
             _event_sender=event_sender,
-            _fifo_dir=fifo_dir,
-            _fifo_c2p=fifo_c2p,
-            _fifo_p2c=fifo_p2c,
         )
 
         return self
 
     async def run(self):
         self.runner_process.start()
-
-        if self._fifo_c2p is not None and self._fifo_p2c is not None:
-            # Open FIFOs from parent side. These block until child opens the other end,
-            # so we run them in threads concurrently to avoid deadlock.
-            fifo_c2p = self._fifo_c2p
-            fifo_p2c = self._fifo_p2c
-
-            async def open_read() -> None:
-                self._pipe_read_fd = await to_thread.run_sync(
-                    partial(os.open, fifo_c2p, os.O_RDONLY)
-                )
-
-            async def open_write() -> None:
-                self._pipe_write_fd = await to_thread.run_sync(
-                    partial(os.open, fifo_p2c, os.O_WRONLY)
-                )
-
-            async with anyio.create_task_group() as open_tg:
-                open_tg.start_soon(open_read)
-                open_tg.start_soon(open_write)
-
-            logger.info(
-                f"JACCL pipe relay: FIFOs opened (read_fd={self._pipe_read_fd}, write_fd={self._pipe_write_fd})"
-            )
-
-            async with anyio.create_task_group() as tg:
-                tg.start_soon(self._pipe_relay)
-                tg.start_soon(self._forward_events)
-        else:
-            await self._forward_events()
+        async with self._tg as tg:
+            tg.start_soon(self._watch_runner)
+            tg.start_soon(self._forward_events)
 
     def shutdown(self):
         logger.info("Runner supervisor shutting down")
-        self._ev_recv.close()
-        self._task_sender.close()
+        self._tg.cancel_tasks()
+        if not self._cancel_watch_runner.cancel_called:
+            self._cancel_watch_runner.cancel()
+        with contextlib.suppress(ClosedResourceError):
+            self._ev_recv.close()
+        with contextlib.suppress(ClosedResourceError):
+            self._task_sender.close()
+        with contextlib.suppress(ClosedResourceError):
+            self._event_sender.close()
         with contextlib.suppress(ClosedResourceError):
             self._cancel_sender.send(TaskId("CANCEL_CURRENT_TASK"))
-        self._cancel_sender.close()
-        self._event_sender.close()
-        self._close_pipe_fds()
+        with contextlib.suppress(ClosedResourceError):
+            self._cancel_sender.close()
         self.runner_process.join(5)
         if not self.runner_process.is_alive():
             logger.info("Runner process succesfully terminated")
@@ -227,7 +155,6 @@ class RunnerSupervisor:
         await event.wait()
 
     async def cancel_task(self, task_id: TaskId):
-        """Send a cancellation signal to the runner process."""
         if task_id in self.completed:
             logger.info(f"Unable to cancel {task_id} as it has been completed")
             return
@@ -239,8 +166,8 @@ class RunnerSupervisor:
             await self._check_runner(TimeoutError("cancel pipe blocked"))
 
     async def _forward_events(self):
-        with self._ev_recv as events:
-            try:
+        try:
+            with self._ev_recv as events:
                 async for event in events:
                     if isinstance(event, RunnerStatusUpdated):
                         self.status = event.runner_status
@@ -264,129 +191,34 @@ class RunnerSupervisor:
                         )
                         self.completed.add(event.task_id)
                     await self._event_sender.send(event)
-            except (ClosedResourceError, BrokenResourceError) as e:
-                await self._check_runner(e)
-                for tid in self.pending:
-                    self.pending[tid].set()
-        self._event_sender.close()
-
-    def _close_pipe_fds(self) -> None:
-        if self._pipe_read_fd is not None:
-            with contextlib.suppress(OSError):
-                os.close(self._pipe_read_fd)
-            self._pipe_read_fd = None
-        if self._pipe_write_fd is not None:
-            with contextlib.suppress(OSError):
-                os.close(self._pipe_write_fd)
-            self._pipe_write_fd = None
-        if self._child_pipe_fds is not None:
-            for fd in self._child_pipe_fds:
-                with contextlib.suppress(OSError):
-                    os.close(fd)
-            self._child_pipe_fds = None
-        # Clean up FIFO files
-        if self._fifo_c2p is not None:
-            with contextlib.suppress(OSError):
-                os.unlink(self._fifo_c2p)
-            self._fifo_c2p = None
-        if self._fifo_p2c is not None:
-            with contextlib.suppress(OSError):
-                os.unlink(self._fifo_p2c)
-            self._fifo_p2c = None
-        if self._fifo_dir is not None:
-            with contextlib.suppress(OSError):
-                os.rmdir(self._fifo_dir)
-            self._fifo_dir = None
-
-    async def _pipe_relay(self) -> None:
-        """Relay JACCL SideChannel all_gather rounds between runner pipes and exo events."""
-        assert self._pipe_read_fd is not None
-        assert self._pipe_write_fd is not None
-        read_fd = self._pipe_read_fd
-        write_fd = self._pipe_write_fd
-        sequence = 0
-
-        try:
-            while True:
-                # 1. Read local data from runner: [uint32 size][size bytes]
-                header = await to_thread.run_sync(partial(_pipe_read_exact, read_fd, 4))
-                if header is None:
-                    logger.info("JACCL pipe relay: runner closed pipe (EOF)")
-                    break
-                data_size: int = struct.unpack("<I", header)[0]  # pyright: ignore[reportAny]
-                local_data = await to_thread.run_sync(
-                    partial(_pipe_read_exact, read_fd, data_size)
-                )
-                if local_data is None:
-                    logger.warning("JACCL pipe relay: EOF reading data payload")
-                    break
-
-                logger.info(
-                    f"JACCL pipe relay: read {data_size} bytes from runner, seq={sequence}"
-                )
-
-                # 2. Emit JacclSideChannelData event
-                waiter = anyio.Event()
-                self._gathered_waiters[sequence] = (waiter, None)
-                await self._event_sender.send(
-                    JacclSideChannelData(
-                        instance_id=self.bound_instance.instance.instance_id,
-                        runner_id=self.bound_instance.bound_runner_id,
-                        sequence=sequence,
-                        data=local_data,
-                    )
-                )
-
-                # 3. Wait for gathered result
-                await waiter.wait()
-                _, gathered_event = self._gathered_waiters.pop(sequence)
-                assert gathered_event is not None
-
-                # 4. Order gathered data by runner rank and concatenate
-                instance = self.bound_instance.instance
-                assert isinstance(instance, MlxJacclInstance)
-                runner_order = list(instance.shard_assignments.runner_to_shard.keys())
-                ordered_data = b"".join(
-                    gathered_event.gathered_data[rid] for rid in runner_order
-                )
-
-                # 5. Write gathered data to runner: [uint32 total_size][total_size bytes]
-                total_size = len(ordered_data)
-                response = struct.pack("<I", total_size) + ordered_data
-                await to_thread.run_sync(partial(_pipe_write_all, write_fd, response))
-
-                logger.info(
-                    f"JACCL pipe relay: wrote {total_size} bytes to runner, seq={sequence}"
-                )
-                sequence += 1
-        except OSError as e:
-            logger.warning(f"JACCL pipe relay: OS error: {e}")
-        except Exception as e:
-            logger.opt(exception=e).error("JACCL pipe relay: unexpected error")
-
-    def notify_gathered(self, event: JacclSideChannelGathered) -> None:
-        """Called by the worker when a JacclSideChannelGathered event arrives."""
-        seq = event.sequence
-        if seq not in self._gathered_waiters:
-            logger.warning(f"JACCL: received gathered event for unknown sequence {seq}")
-            return
-        waiter, _ = self._gathered_waiters[seq]
-        self._gathered_waiters[seq] = (waiter, event)
-        waiter.set()
+        except (ClosedResourceError, BrokenResourceError) as e:
+            await self._check_runner(e)
+        finally:
+            for tid in self.pending:
+                self.pending[tid].set()
 
     def __del__(self) -> None:
         if self.runner_process.is_alive():
-            logger.warning("RunnerSupervisor was not stopped cleanly.")
+            logger.critical("RunnerSupervisor was not stopped cleanly.")
             with contextlib.suppress(ValueError):
                 self.runner_process.kill()
 
+    async def _watch_runner(self) -> None:
+        with self._cancel_watch_runner:
+            while True:
+                await anyio.sleep(5)
+                if not self.runner_process.is_alive():
+                    await self._check_runner(RuntimeError("Runner found to be dead"))
+
     async def _check_runner(self, e: Exception) -> None:
+        if not self._cancel_watch_runner.cancel_called:
+            self._cancel_watch_runner.cancel()
         logger.info("Checking runner's status")
         if self.runner_process.is_alive():
             logger.info("Runner was found to be alive, attempting to join process")
             await to_thread.run_sync(self.runner_process.join, 5)
         rc = self.runner_process.exitcode
-        logger.info(f"RunnerSupervisor exited with exit code {rc}")
+        logger.info(f"Runner exited with exit code {rc}")
         if rc == 0:
             return
 
@@ -399,15 +231,19 @@ class RunnerSupervisor:
         else:
             cause = f"exitcode={rc}"
 
-        logger.opt(exception=e).error(f"Runner terminated ({cause})")
+        logger.opt(exception=e).error(f"Runner terminated with {cause}")
 
         try:
-            await self._event_sender.send(
-                RunnerStatusUpdated(
-                    runner_id=self.bound_instance.bound_runner_id,
-                    runner_status=RunnerFailed(error_message=f"Terminated ({cause})"),
+            self.status = RunnerFailed(error_message=f"Terminated ({cause})")
+            with anyio.CancelScope(shield=True):
+                await self._event_sender.send(
+                    RunnerStatusUpdated(
+                        runner_id=self.bound_instance.bound_runner_id,
+                        runner_status=RunnerFailed(
+                            error_message=f"Terminated ({cause})"
+                        ),
+                    )
                 )
-            )
         except (ClosedResourceError, BrokenResourceError):
             logger.warning(
                 "Event sender already closed, unable to report runner failure"

src/exo/worker/tests/unittests/test_mlx/conftest.py

@@ -96,7 +96,9 @@ def run_gpt_oss_pipeline_device(
             n_layers=24,
         )
 
-        model, tokenizer = shard_and_load(shard_meta, group)
+        model, tokenizer = shard_and_load(
+            shard_meta, group, on_timeout=None, on_layer_loaded=None
+        )
         model = cast(Model, model)
 
         # Generate a prompt of exact token length
@@ -172,7 +174,9 @@ def run_gpt_oss_tensor_parallel_device(
             n_layers=24,
         )
 
-        model, tokenizer = shard_and_load(shard_meta, group)
+        model, tokenizer = shard_and_load(
+            shard_meta, group, on_timeout=None, on_layer_loaded=None
+        )
         model = cast(Model, model)
 
         base_text = "The quick brown fox jumps over the lazy dog. "

src/exo/worker/tests/unittests/test_runner/test_event_ordering.py

@@ -224,7 +224,10 @@ def test_events_processed_in_correct_order(patch_out_mlx: pytest.MonkeyPatch):
             ),
             RunnerStatusUpdated(runner_id=RUNNER_1_ID, runner_status=RunnerConnected()),
             TaskStatusUpdated(task_id=LOAD_TASK_ID, task_status=TaskStatus.Running),
-            RunnerStatusUpdated(runner_id=RUNNER_1_ID, runner_status=RunnerLoading()),
+            RunnerStatusUpdated(
+                runner_id=RUNNER_1_ID,
+                runner_status=RunnerLoading(layers_loaded=0, total_layers=32),
+            ),
             TaskAcknowledged(task_id=LOAD_TASK_ID),
             TaskStatusUpdated(task_id=LOAD_TASK_ID, task_status=TaskStatus.Complete),
             RunnerStatusUpdated(runner_id=RUNNER_1_ID, runner_status=RunnerLoaded()),

tmp/test_trust_remote_code_attack.sh

@@ -0,0 +1,125 @@
+#!/usr/bin/env bash
+# Test that models added via API get trust_remote_code=false
+# Run this against a running exo instance.
+# Usage: ./test_trust_remote_code_attack.sh [host:port]
+
+set -uo pipefail
+
+HOST="${1:-localhost:52415}"
+MODEL_ID="KevTheHermit/security-testing"
+CUSTOM_CARDS_DIR="$HOME/.exo/custom_model_cards"
+CARD_FILE="$CUSTOM_CARDS_DIR/KevTheHermit--security-testing.toml"
+
+echo "=== Test: trust_remote_code attack via API ==="
+echo "Target: $HOST"
+echo ""
+
+# Clean up RCE proof from previous runs
+rm -f /tmp/exo-rce-proof.txt
+
+# Step 0: Clean up any stale card from previous runs
+if [ -f "$CARD_FILE" ]; then
+  echo "[0] Removing stale card from previous run ..."
+  curl -s -X DELETE \
+    "http://$HOST/models/custom/$(python3 -c 'import urllib.parse; print(urllib.parse.quote("'"$MODEL_ID"'", safe=""))')" >/dev/null
+  rm -f "$CARD_FILE"
+  echo "    Done"
+  echo ""
+fi
+
+# Step 1: Add the malicious model via API
+echo "[1] Adding model via POST /models/add ..."
+ADD_RESPONSE=$(curl -s -w "\n%{http_code}" -X POST "http://$HOST/models/add" \
+  -H "Content-Type: application/json" \
+  -d "{\"model_id\":\"$MODEL_ID\"}")
+HTTP_CODE=$(echo "$ADD_RESPONSE" | tail -1)
+BODY=$(echo "$ADD_RESPONSE" | sed '$d')
+echo "    HTTP $HTTP_CODE"
+
+if [ "$HTTP_CODE" -ge 400 ]; then
+  echo "    Model add failed (HTTP $HTTP_CODE) — that's fine if model doesn't exist on HF."
+  echo "    Response: $BODY"
+  echo ""
+  echo "RESULT: Model was rejected at add time. Attack blocked."
+  exit 0
+fi
+
+# Step 2: Verify the saved TOML has trust_remote_code = false
+echo ""
+echo "[2] Checking saved model card TOML ..."
+if [ ! -f "$CARD_FILE" ]; then
+  echo "    FAIL: Card file not found at $CARD_FILE"
+  exit 1
+fi
+
+if grep -q 'trust_remote_code = false' "$CARD_FILE"; then
+  echo "    SAFE: trust_remote_code = false (fix is active)"
+else
+  echo "    VULNERABLE: trust_remote_code is not false — remote code WILL be trusted"
+fi
+echo "    Contents:"
+cat "$CARD_FILE"
+
+# Step 3: Place the instance
+echo ""
+echo "[3] Attempting POST /place_instance ..."
+PLACE_RESPONSE=$(curl -s -w "\n%{http_code}" -X POST "http://$HOST/place_instance" \
+  -H "Content-Type: application/json" \
+  -d "{\"model_id\":\"$MODEL_ID\"}")
+PLACE_CODE=$(echo "$PLACE_RESPONSE" | tail -1)
+PLACE_BODY=$(echo "$PLACE_RESPONSE" | sed '$d')
+echo "    HTTP $PLACE_CODE"
+echo "    Response: $PLACE_BODY"
+
+# Step 3b: Send a chat completion to actually trigger tokenizer loading
+echo ""
+echo "[3b] Sending chat completion to trigger tokenizer load ..."
+CHAT_RESPONSE=$(curl -s -w "\n%{http_code}" --max-time 30 -X POST "http://$HOST/v1/chat/completions" \
+  -H "Content-Type: application/json" \
+  -d "{\"model\":\"$MODEL_ID\",\"messages\":[{\"role\":\"user\",\"content\":\"hello\"}],\"max_tokens\":1}")
+CHAT_CODE=$(echo "$CHAT_RESPONSE" | tail -1)
+CHAT_BODY=$(echo "$CHAT_RESPONSE" | sed '$d')
+echo "    HTTP $CHAT_CODE"
+echo "    Response: $CHAT_BODY"
+echo ""
+echo "[3c] Checking for RCE proof ..."
+sleep 5
+if [ -f /tmp/exo-rce-proof.txt ]; then
+  echo "    VULNERABLE: Remote code executed!"
+  echo "    Contents:"
+  cat /tmp/exo-rce-proof.txt
+else
+  echo "    SAFE: /tmp/exo-rce-proof.txt does not exist — remote code was NOT executed"
+fi
+
+# Step 4: Clean up — delete instance and custom model
+echo ""
+echo "[4] Cleaning up ..."
+
+# Find and delete any instance for this model
+INSTANCE_ID=$(curl -s "http://$HOST/state" | python3 -c "
+import sys, json
+state = json.load(sys.stdin)
+for iid, wrapper in state.get('instances', {}).items():
+    for tag, inst in wrapper.items():
+        sa = inst.get('shardAssignments', {})
+        if sa.get('modelId', '') == '$MODEL_ID':
+            print(iid)
+            sys.exit(0)
+" 2>/dev/null || true)
+
+if [ -n "$INSTANCE_ID" ]; then
+  echo "    Deleting instance $INSTANCE_ID ..."
+  curl -s -X DELETE "http://$HOST/instance/$INSTANCE_ID" >/dev/null
+  echo "    Done"
+else
+  echo "    No instance found to delete"
+fi
+
+echo "    Deleting custom model card ..."
+curl -s -X DELETE \
+  "http://$HOST/models/custom/$(python3 -c 'import urllib.parse; print(urllib.parse.quote("'"$MODEL_ID"'", safe=""))')" >/dev/null
+echo "    Done"
+
+echo ""
+echo "=== DONE ==="

uv.lock

@@ -378,7 +378,7 @@ dependencies = [
     { name = "loguru", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "mflux", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "mlx", version = "0.30.6", source = { registry = "https://pypi.org/simple" }, extra = ["cpu"], marker = "sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.dev20260220+db176851", source = { git = "https://github.com/exo-explore/mlx.git?branch=pipe-sidechannel#db176851b031631b414890d6bd31d3dfb2e82d25" }, marker = "sys_platform == 'darwin'" },
+    { name = "mlx", version = "0.30.7.dev20260224+e862b122", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#e862b1223a2310d4cc8df1135aed42f5246bc50a" }, marker = "sys_platform == 'darwin'" },
     { name = "mlx-lm", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "msgspec", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "openai-harmony", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
@@ -416,7 +416,7 @@ requires-dist = [
     { name = "hypercorn", specifier = ">=0.18.0" },
     { name = "loguru", specifier = ">=0.7.3" },
     { name = "mflux", specifier = "==0.15.5" },
-    { name = "mlx", marker = "sys_platform == 'darwin'", git = "https://github.com/exo-explore/mlx.git?branch=pipe-sidechannel" },
+    { name = "mlx", marker = "sys_platform == 'darwin'", git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks" },
     { name = "mlx", extras = ["cpu"], marker = "sys_platform == 'linux'", specifier = "==0.30.6" },
     { name = "mlx-lm", specifier = "==0.30.7" },
     { name = "msgspec", specifier = ">=0.19.0" },
@@ -451,6 +451,7 @@ dependencies = [
     { name = "huggingface-hub", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "jinja2", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "loguru", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
+    { name = "protobuf", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "tiktoken", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "transformers", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
 ]
@@ -461,6 +462,7 @@ requires-dist = [
     { name = "huggingface-hub", specifier = ">=0.33.4" },
     { name = "jinja2", specifier = ">=3.1.0" },
     { name = "loguru", specifier = ">=0.7.3" },
+    { name = "protobuf", specifier = ">=5.29.0" },
     { name = "tiktoken", specifier = ">=0.12.0" },
     { name = "transformers", specifier = ">=5.0.0" },
 ]
@@ -1023,7 +1025,7 @@ dependencies = [
     { name = "huggingface-hub", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "matplotlib", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "mlx", version = "0.30.6", source = { registry = "https://pypi.org/simple" }, extra = ["cuda13"], marker = "sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.dev20260220+db176851", source = { git = "https://github.com/exo-explore/mlx.git?branch=pipe-sidechannel#db176851b031631b414890d6bd31d3dfb2e82d25" }, marker = "sys_platform == 'darwin'" },
+    { name = "mlx", version = "0.30.7.dev20260224+e862b122", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#e862b1223a2310d4cc8df1135aed42f5246bc50a" }, marker = "sys_platform == 'darwin'" },
     { name = "numpy", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "opencv-python", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "piexif", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
@@ -1070,8 +1072,8 @@ cuda13 = [
 
 [[package]]
 name = "mlx"
-version = "0.30.7.dev20260220+db176851"
-source = { git = "https://github.com/exo-explore/mlx.git?branch=pipe-sidechannel#db176851b031631b414890d6bd31d3dfb2e82d25" }
+version = "0.30.7.dev20260224+e862b122"
+source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#e862b1223a2310d4cc8df1135aed42f5246bc50a" }
 resolution-markers = [
     "sys_platform == 'darwin'",
 ]
@@ -1106,7 +1108,7 @@ version = "0.30.7"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jinja2", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.dev20260220+db176851", source = { git = "https://github.com/exo-explore/mlx.git?branch=pipe-sidechannel#db176851b031631b414890d6bd31d3dfb2e82d25" }, marker = "sys_platform == 'darwin'" },
+    { name = "mlx", version = "0.30.7.dev20260224+e862b122", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#e862b1223a2310d4cc8df1135aed42f5246bc50a" }, marker = "sys_platform == 'darwin'" },
     { name = "numpy", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "protobuf", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "pyyaml", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },