electron-rare/exo
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • 5c3f0e3ac6 faster initial node discovery Alex Cheema 2024-07-17 17:11:50 -07:00
  • 8a35fd83f6 support chatgpt api endpoint fron any node #24 Alex Cheema 2024-07-17 15:37:47 -07:00
  • ba7abb9896 fix ring topology img Alex Cheema 2024-07-17 15:14:01 -07:00
  • c432871ef5 replace the ring topology image as it was not rendering sometimes Alex Cheema 2024-07-17 15:10:31 -07:00
  • bcab97cb41 instructions for how to force a node to be the tail Alex Cheema 2024-07-17 15:00:23 -07:00
  • eb92da2c3e cleaner chatgpt api impl with async callbacks Alex Cheema 2024-07-17 14:50:16 -07:00
  • 7c97ef521d DEBUG should be imported from exo Alex Cheema 2024-07-17 14:24:55 -07:00
  • 5055e3788d separate prerequisities seciton / troubleshooting section in installation of readme Alex Cheema 2024-07-17 11:54:09 -07:00
  • 998d484384 match psutil platform detection might catch some edge cases Alex Cheema 2024-07-17 11:48:52 -07:00
  • 99d40b1df0 download tinygrad model log Alex Cheema 2024-07-17 11:47:47 -07:00
  • 35335706d6 Merge pull request #21 from the-alex-b/setup Alex Cheema 2024-07-17 11:14:31 -07:00
  • 442c7d8c3e update readme Alex 2024-07-17 13:22:29 +02:00
  • 879b969028 Removed requirements.txt Alex 2024-07-17 13:22:12 +02:00
  • 4ea7db854a more complete python gitignore template Alex 2024-07-17 13:22:04 +02:00
  • bfaeccc7d5 added setup py Alex 2024-07-17 13:21:46 +02:00
  • 12fcbc0d4c switch over to psutil, more robust system detection Alex Cheema 2024-07-17 00:53:46 -07:00
  • 43fa45990e Merge pull request #18 from andvarfolomeev/fix/requirements Alex Cheema 2024-07-16 23:21:05 -07:00
  • 7545e0605b fix: syntax error in requirements.txt Andrey Varfolomeev 2024-07-17 16:12:38 +10:00
  • 6b3727f023 fetch model if doesnt exist on tinygrad Alex Cheema 2024-07-16 23:09:42 -07:00
  • b1f3204e80 add Jinja requirement for linux Alex Cheema 2024-07-16 22:56:40 -07:00
  • 365114ec57 add stars to readme Alex Cheema 2024-07-16 22:46:09 -07:00
  • 71e00745cc fix tokenizer inconsistencies Alex Cheema 2024-07-16 22:44:50 -07:00
  • c673b4c37b clarify readme Alex Cheema 2024-07-16 21:21:49 -07:00
  • 7cb1ba55a4 Clarify readme iOS Alex Cheema 2024-07-16 21:19:09 -07:00
  • c819f675cb fix linux/amd gpu memory, convert to MB Alex Cheema 2024-07-16 18:38:57 -07:00
  • e93a319ca4 typo Alex Cheema 2024-07-16 18:34:13 -07:00
  • ce46f00059 linux device capabilities Alex Cheema 2024-07-16 18:33:29 -07:00
  • dbbc7be57f remove hard dependency on MLX fixes #8 Alex Cheema 2024-07-16 18:07:45 -07:00
  • 5e8bfc8a90 remove deprecated main_static Alex Cheema 2024-07-16 17:57:01 -07:00
  • e049f701ce debugging instructions in README Alex Cheema 2024-07-16 16:59:49 -07:00
  • bde1e53f5f add license Alex Cheema 2024-07-16 14:06:36 -07:00
  • dd8d18128c add an opaque inference_state that inference engines can use to pass around small state to other devices Alex Cheema 2024-07-16 13:05:12 -07:00
  • 03ba31c020 wip state Alex Cheema 2024-07-16 10:13:13 -07:00
  • ed7672e32c Update README.md Alex Cheema 2024-07-16 08:39:57 -07:00
  • 9354081b21 Merge pull request #9 from mattroyer/patch-1 Alex Cheema 2024-07-16 07:59:31 -07:00
  • 9324fc15d9 Fix broken links Matt Royer 2024-07-16 07:47:59 -06:00
  • b897fa4420 Typo ring memory weighted partitioning strategy Alex Cheema 2024-07-16 02:42:02 -07:00
  • 50d5e948f8 Update README.md Alex Cheema 2024-07-16 02:40:50 -07:00
  • b6d91972fc add notice of python>=3.12.0 Alex Cheema 2024-07-16 02:11:24 -07:00
  • 403abcfa41 smaller ring topology img Alex Cheema 2024-07-16 02:06:28 -07:00
  • 231cde5ff5 ring topology image Alex Cheema 2024-07-16 02:05:58 -07:00
  • d78d5b201a explain device equality in README Alex Cheema 2024-07-16 02:05:50 -07:00
  • 94b6a2494b print debug only Alex Cheema 2024-07-16 01:28:30 -07:00
  • bf565f945d fix #7 no module named aiohttp Alex Cheema 2024-07-16 01:09:39 -07:00
  • bdf105a60c clarify example in readme Alex Cheema 2024-07-16 00:32:40 -07:00
  • 9759408a67 trim off the eos_token_id from chatgpt api response Alex Cheema 2024-07-16 00:21:17 -07:00
  • f2895cbcee revive the chatgpt api endpoint on :8000 Alex Cheema 2024-07-16 00:17:23 -07:00
  • 1d5c28aed4 (partially) restore exo node equality by forwarding prompts to the dynamically selected head Alex Cheema 2024-07-15 22:59:00 -07:00
  • 1ec92b731e README notice about api endpoint Alex Cheema 2024-07-15 22:46:06 -07:00
  • 108a904ab3 Readme whitespace Alex Cheema 2024-07-15 22:36:07 -07:00
  • e17905e295 add global reset Alex Cheema 2024-07-15 22:35:54 -07:00
  • de9b89ea29 readme typo Alex Cheema 2024-07-15 18:22:42 -07:00
  • 199eeb03db known issues section in readme Alex Cheema 2024-07-15 18:22:09 -07:00
  • d2184f583a keep track of already visited peers in global operations: collect_topology Alex Cheema 2024-07-15 15:43:00 -07:00
  • 4502da5bc4 readme bug notice Alex Cheema 2024-07-15 15:23:58 -07:00
  • f9a201ddbf docs dir Alex Cheema 2024-07-15 11:37:22 -07:00
  • 4d43cb9174 logo Alex Cheema 2024-07-15 11:37:12 -07:00
  • 4e1f01eeec update links Alex Cheema 2024-07-15 11:20:42 -07:00
  • 1dea8b9c28 update discord link Alex Cheema 2024-07-15 11:15:06 -07:00
  • 544c229e8a discord link Alex Cheema 2024-07-15 11:02:13 -07:00
  • 98b30e056b readme tweak Alex Cheema 2024-07-15 00:08:42 -07:00
  • 963f8eb6a1 better logs for DEBUG>=1 Alex Cheema 2024-07-14 23:55:27 -07:00
  • a009f7d608 move examples to examples dir Alex Cheema 2024-07-14 23:48:20 -07:00
  • b6595bac04 add llama-3-70b to the examples Alex Cheema 2024-07-14 23:47:33 -07:00
  • 54e8cad2d6 remove uneeded prints Alex Cheema 2024-07-14 23:31:18 -07:00
  • c691205591 empty space Alex Cheema 2024-07-14 23:29:20 -07:00
  • bcd58938de clean debug logs Alex Cheema 2024-07-14 23:28:55 -07:00
  • b9c323bb07 memory-efficient shard loading Alex Cheema 2024-07-14 23:27:57 -07:00
  • 53a5b3fc6a add uuid requirement Alex Cheema 2024-07-14 21:46:41 -07:00
  • 05b9fa497d initialize node id to uuid4 if not set Alex Cheema 2024-07-14 21:46:30 -07:00
  • ff597d9551 fix discovery Alex Cheema 2024-07-14 21:46:13 -07:00
  • a04974168e fix model import path Alex Cheema 2024-07-14 21:46:00 -07:00
  • b8a2a0fbe0 update readme run instruction Alex Cheema 2024-07-14 21:26:56 -07:00
  • a933352ac3 add DEBUG flag for controlling debug logs Alex Cheema 2024-07-14 21:26:45 -07:00
  • dd882fe6bc experimental notice Alex Cheema 2024-07-14 21:16:20 -07:00
  • c8753ba5fe reshuffle readme Alex Cheema 2024-07-14 21:12:55 -07:00
  • ee5204fbca readme installation instructions Alex Cheema 2024-07-14 21:12:17 -07:00
  • 78da11e10b slightly nicer readme Alex Cheema 2024-07-14 21:05:41 -07:00
  • 2fc472c8fe slightly nicer readme Alex Cheema 2024-07-14 21:03:48 -07:00
  • 8ff3e263a0 slightly nicer readme Alex Cheema 2024-07-14 21:02:25 -07:00
  • 32f2e36fd3 main rename Alex Cheema 2024-07-14 21:01:28 -07:00
  • 5bbde22a23 move everything under exo module Alex Cheema 2024-07-14 21:00:37 -07:00
  • c851644a43 update requirements, specify exact versions Alex Cheema 2024-07-14 20:55:29 -07:00
  • 32972033dd update readme Alex Cheema 2024-07-14 18:38:48 -07:00
  • 5ef07d41a5 readme Alex Cheema 2024-07-14 18:09:38 -07:00
  • 490fa102a4 tinygrad inference engine Alex Cheema 2024-07-14 13:07:37 -07:00
  • e6f387a690 handle is_finished Alex Cheema 2024-07-13 23:27:34 -07:00
  • b01f69bb6b add support for multiple concurrent requests with request ids Alex Cheema 2024-07-13 23:11:01 -07:00
  • 7077652c8e graceful node shutdown Alex Cheema 2024-07-13 20:43:37 -07:00
  • ca6095c04d a generic test for every inference engine Alex Cheema 2024-07-13 18:25:26 -07:00
  • 850b72d3ea make StatefulShardedModel callable, add some tests for mlx sharded inference Alex Cheema 2024-07-13 15:41:15 -07:00
  • 6ee0547eff fix layer calculation for sharded llama Alex Cheema 2024-07-13 15:39:31 -07:00
  • 445eda156c dynamically assign shards to nodes deterministically weighted by memory Alex Cheema 2024-06-25 21:17:58 +01:00
  • 36b8456798 collect global topology with local peer visibility, ring memory weighted partitioning strategy Alex Cheema 2024-06-25 12:32:03 +01:00
  • 3a66a0a4a8 add requirements.txt Alex Cheema 2024-06-24 21:00:04 +01:00
  • ee96c6b023 add another test for device capabiities on MacBook Air Alex Cheema 2024-06-24 20:59:55 +01:00
  • 6c8c9ee7b1 topology with partitioning strategy Alex Cheema 2024-06-24 20:56:50 +01:00
  • 563dcb56b0 mlx sharded implementation with example of distributed inference Alex Cheema 2024-06-24 19:35:57 +01:00
  • a21f59ff45 scaffolding for networking, inference and orchestration Alex Cheema 2024-06-23 23:28:10 +01:00