L'électron rareandClaude Opus 4.6
0e36e71183
Lot 401: Auto-free ComfyUI VRAM every 5min (cron) Lot 402: /benchmark — test all services latency in parallel Lot 403: Ctrl+F chat search overlay with highlight Lot 404: /translate (/tr) — FR↔EN auto-detect via Ollama Lot 405: Persona mood system (time-of-day personality modifier) 110 commands, 211 tests. Co-Authored-By: Claude Opus 4.6 (1M context) <[email protected]>
99 lines
3.9 KiB
Bash
Executable File
99 lines
3.9 KiB
Bash
Executable File
#!/bin/bash
|
|
# =============================================================================
|
|
# KXKM-AI Optimization Script — requires sudo
|
|
# Run on kxkm-ai: sudo bash scripts/sudo-optimize.sh
|
|
# =============================================================================
|
|
|
|
set -e
|
|
echo "╔══════════════════════════════════════╗"
|
|
echo "║ KXKM-AI Optimization (sudo) ║"
|
|
echo "╚══════════════════════════════════════╝"
|
|
echo ""
|
|
|
|
# --- 1. Ollama tuning ---
|
|
echo "▸ [1/4] Ollama tuning..."
|
|
mkdir -p /etc/systemd/system/ollama.service.d
|
|
cat > /etc/systemd/system/ollama.service.d/override.conf << 'EOF'
|
|
[Service]
|
|
Environment="OLLAMA_NUM_PARALLEL=2"
|
|
Environment="OLLAMA_FLASH_ATTENTION=1"
|
|
Environment="OLLAMA_KV_CACHE_TYPE=q8_0"
|
|
Environment="OLLAMA_MAX_LOADED_MODELS=2"
|
|
Environment="OLLAMA_ORIGINS=*"
|
|
EOF
|
|
echo " Written /etc/systemd/system/ollama.service.d/override.conf"
|
|
systemctl daemon-reload
|
|
systemctl restart ollama
|
|
echo " Ollama restarted with: NUM_PARALLEL=2 FLASH_ATTENTION=1 KV_CACHE=q8_0"
|
|
sleep 3
|
|
|
|
# --- 2. Docker GPU runtime ---
|
|
echo ""
|
|
echo "▸ [2/4] Docker GPU runtime..."
|
|
if nvidia-ctk --version > /dev/null 2>&1; then
|
|
echo " nvidia-container-toolkit already installed"
|
|
else
|
|
echo " Installing nvidia-container-toolkit..."
|
|
apt-get update -qq && apt-get install -y -qq nvidia-container-toolkit
|
|
nvidia-ctk runtime configure --runtime=docker
|
|
systemctl restart docker
|
|
echo " Docker GPU runtime configured"
|
|
fi
|
|
|
|
# --- 3. Linger for kxkm user services ---
|
|
echo ""
|
|
echo "▸ [3/4] Enabling linger for kxkm..."
|
|
loginctl enable-linger kxkm 2>/dev/null && echo " Linger enabled" || echo " Linger already enabled"
|
|
|
|
# --- 4. Free ComfyUI VRAM ---
|
|
echo ""
|
|
echo "▸ [4/4] Freeing ComfyUI VRAM..."
|
|
curl -s -X POST http://localhost:8188/free -H 'Content-Type: application/json' -d '{"unload_models":true,"free_memory":true}' > /dev/null 2>&1 && echo " ComfyUI VRAM freed" || echo " ComfyUI not reachable (skip)"
|
|
|
|
# --- Verify ---
|
|
echo ""
|
|
echo "════════════════════════════════════════"
|
|
echo " Verification"
|
|
echo "════════════════════════════════════════"
|
|
|
|
# Ollama
|
|
echo ""
|
|
echo "▸ Ollama:"
|
|
systemctl is-active ollama && echo " Status: running" || echo " Status: STOPPED"
|
|
sleep 2
|
|
curl -s http://localhost:11434/api/ps | python3 -c "
|
|
import sys,json
|
|
d=json.load(sys.stdin)
|
|
for m in d.get('models',[]):
|
|
print(f' Model: {m[\"name\"]} ({m.get(\"size_vram\",0)/(1024**3):.1f}GB VRAM)')
|
|
if not d.get('models'):
|
|
print(' No models loaded (will load on first request)')
|
|
" 2>/dev/null || echo " Ollama not responding yet"
|
|
|
|
# GPU
|
|
echo ""
|
|
echo "▸ GPU:"
|
|
nvidia-smi --query-gpu=memory.used,memory.free,memory.total --format=csv,noheader 2>/dev/null | while read line; do
|
|
echo " VRAM: $line"
|
|
done
|
|
|
|
# Docker
|
|
echo ""
|
|
echo "▸ Docker GPU:"
|
|
docker info 2>/dev/null | grep -i "runtimes\|nvidia" | head -3 | while read line; do
|
|
echo " $line"
|
|
done
|
|
|
|
# Services
|
|
echo ""
|
|
echo "▸ User services (kxkm):"
|
|
su - kxkm -c "systemctl --user status kokoro-tts --no-pager 2>/dev/null | head -3" 2>/dev/null
|
|
su - kxkm -c "systemctl --user status ai-bridge --no-pager 2>/dev/null | head -3" 2>/dev/null
|
|
|
|
echo ""
|
|
echo "════════════════════════════════════════"
|
|
echo " Done. Reload Ollama model:"
|
|
echo " curl -s http://localhost:11434/api/chat \\"
|
|
echo " -d '{\"model\":\"qwen3.5:9b\",\"messages\":[{\"role\":\"user\",\"content\":\"ping\"}],\"stream\":false,\"options\":{\"num_predict\":1},\"keep_alive\":\"30m\"}'"
|
|
echo "════════════════════════════════════════"
|