Commit Graph

  • 4937fb3df8 Merge pull request #597 from exo-explore/tuioverflow Alex Cheema 2025-01-08 16:40:16 +00:00
  • 2d631ea53d Merge pull request #596 from exo-explore/phi4 Alex Cheema 2025-01-08 16:39:32 +00:00
  • 2846a9122f tok tests phi4 Alex Cheema 2025-01-08 16:39:11 +00:00
  • 553ccce728 fix prompt and output overflow in tui tuioverflow Alex Cheema 2025-01-08 16:36:56 +00:00
  • c587593364 add phi 3.5, phi 4 Alex Cheema 2025-01-08 16:19:43 +00:00
  • 3c9efe103d Merge pull request #590 from metaspartan/fix-models-api v0.0.4-alpha Alex Cheema 2025-01-07 02:32:06 +00:00
  • 627bfcae7c Fix the /v1/models API to output proper OpenAI compatible endpoint Carsen Klock 2025-01-06 01:20:30 -07:00
  • d9a836f152 Merge pull request #588 from exo-explore/betterdl v0.0.3-alpha Alex Cheema 2025-01-05 02:35:04 +00:00
  • 29244c6369 fix args for ensure_shard betterdl Alex Cheema 2025-01-05 02:33:25 +00:00
  • 8c191050a2 download status in parallel, support async ensure shard with using shard_downloader instead Alex Cheema 2025-01-05 02:31:59 +00:00
  • 7b1656140e Merge pull request #585 from pepebruari/main Alex Cheema 2025-01-03 23:49:50 +00:00
  • fe50d4d34d Add --system-prompt to exo cli pepebruari 2025-01-03 16:16:22 -05:00
  • 03aa6cecf1 Merge pull request #584 from exo-explore/AlexCheema-patch-1 Alex Cheema 2024-12-31 17:51:10 +00:00
  • 178cc4d961 add trending badge to README.md AlexCheema-patch-1 Alex Cheema 2024-12-31 17:50:29 +00:00
  • b13e368368 fix inference engine Pranav Veldurthi 2024-12-30 19:41:19 -05:00
  • 9986fb86d4 remove prints and fix download progress for SD Pranav Veldurthi 2024-12-30 19:07:37 -05:00
  • 3475be9e9e Remove build Pranav Veldurthi 2024-12-30 18:39:17 -05:00
  • fff8a1a690 fix inference engine for inference state Pranav Veldurthi 2024-12-30 18:36:53 -05:00
  • 54605299b8 Merge Latest Pranav Veldurthi 2024-12-30 18:36:23 -05:00
  • a174c78004 Merge pull request #383 from ianpaul10/feat/manual-disc-follow-up v0.0.2-alpha v0.0.1-alpha Alex Cheema 2024-12-28 11:57:25 +00:00
  • b003292b89 formatting and fixing tests after rebasing Ian Paul 2024-12-28 12:31:15 +07:00
  • 1dfd058c23 rm unecessary lock Ian Paul 2024-11-28 15:22:45 +07:00
  • 2eadaa2c0d rm redundant cleanup task Ian Paul 2024-11-25 18:32:34 +07:00
  • 637446ffa9 rm redundant typing Ian Paul 2024-11-15 11:03:35 +07:00
  • a31f9e6c20 fix test warnings Ian Paul 2024-11-15 10:52:00 +07:00
  • 18acb97b42 make popping from dict threadsafe Ian Paul 2024-11-15 10:39:21 +07:00
  • b066c944f3 make all I/O ops in manual_discovery.py run inside a ThreadPoolExecutor Ian Paul 2024-11-15 10:20:42 +07:00
  • 0e34ce2169 patch after rebasing to main Ian Paul 2024-11-06 11:40:03 +07:00
  • 90de7eada9 changes after rebase Ian Paul 2024-11-06 10:50:58 +07:00
  • 8d24df2b4b fix test runtime warning Ian Paul 2024-10-24 17:03:50 +07:00
  • e5eb3259a5 handle when a peer is removed from config, so the known_peers dict gets updated accordingly Ian Paul 2024-10-24 09:43:20 +07:00
  • 2e8227fccb handle intermediate state for when config is being updated Ian Paul 2024-10-24 09:16:46 +07:00
  • 98118babae allow update to manual discovery file Ian Paul 2024-10-24 09:10:10 +07:00
  • 496a3b49f5 Merge pull request #561 from VerisimilitudeX/patch-1 Alex Cheema 2024-12-27 17:06:00 +00:00
  • aba1bed5ed Merge pull request #575 from exo-explore/fixtok Alex Cheema 2024-12-27 16:36:34 +00:00
  • e08522ee97 Revert "Merge pull request #573 from damho1104/feature/add-exaone-3.5-model" fixtok Alex Cheema 2024-12-27 16:35:54 +00:00
  • 4eb6a6a74a Merge pull request #573 from damho1104/feature/add-exaone-3.5-model Alex Cheema 2024-12-27 12:36:09 +00:00
  • 94a5e908b0 add exaone-3.5 LLM Model damho.lee 2024-12-24 20:57:11 +09:00
  • fdc3b5ac02 Merge pull request #571 from exo-explore/function_calling Alex Cheema 2024-12-24 02:08:48 +00:00
  • 185b1e375c fix names in dummy tokenizer function_calling Alex Cheema 2024-12-24 02:08:20 +00:00
  • 078b807654 fix names of qwen models Alex Cheema 2024-12-24 02:06:13 +00:00
  • 188ac445c9 function calling example with weather tool Alex Cheema 2024-12-24 01:57:17 +00:00
  • 456fbdd2b0 add chatgpt-api-compatible tools for function calling Alex Cheema 2024-12-24 01:51:55 +00:00
  • 41df9ce1d7 Merge pull request #570 from exo-explore/moreqwen Alex Cheema 2024-12-24 01:51:26 +00:00
  • c609c05e40 add qwen-2.5-1.5b, qwen-2.5-3b, qwen-2.5-32b moreqwen Alex Cheema 2024-12-24 01:50:12 +00:00
  • ba8c514974 Merge pull request #569 from deftdawg/env_bash Alex Cheema 2024-12-22 23:25:38 +00:00
  • cde912deef - Use #!/usr/bin/env bash instead of #!/bin/bash for better portability DeftDawg 2024-12-22 01:14:54 -05:00
  • 154e0f58e4 Implement suggestiond Piyush Acharya 2024-12-21 19:40:53 -08:00
  • c609f39996 disable the m3 max 128GB test runners Alex Cheema 2024-12-19 01:24:47 +00:00
  • 0278de7b7e noop tracing Alex Cheema 2024-12-18 23:32:25 +00:00
  • b02c0a5be0 new approach to mlx async operations and make tokenizer operations async too Alex Cheema 2024-12-18 23:26:42 +00:00
  • 6c82365ee2 Improved clarity, fixed typos, added macOS/Linux examples, and enhanced installation/debugging instructions Piyush Acharya 2024-12-17 18:02:34 -08:00
  • 165a9e1f53 more granular tracing Alex Cheema 2024-12-18 00:13:35 +00:00
  • db010d51fb distributed tracing Alex Cheema 2024-12-18 00:01:14 +00:00
  • 023ddc207e support different network interface tests Alex Cheema 2024-12-17 21:03:00 +00:00
  • 2f0b543a1e add peer connection info to tinychat Alex Cheema 2024-12-17 17:37:40 +00:00
  • 7ac4004392 change it back to collecting topology periodically even if peers dont change Alex Cheema 2024-12-17 17:32:18 +00:00
  • 198308b1eb more robust udp broadcast Alex Cheema 2024-12-17 17:28:55 +00:00
  • 1f108a06ff remove test sleep Alex Cheema 2024-12-17 16:47:05 +00:00
  • 3a58576f8c make sure this is actually doing something Alex Cheema 2024-12-17 16:22:22 +00:00
  • 0a07223074 switch to uvloop (faster asyncio event loop) and optimise grpc settings Alex Cheema 2024-12-17 16:10:56 +00:00
  • 58f0a0f547 optimise grpc parameters Alex Cheema 2024-12-17 14:50:52 +00:00
  • 5c0cd1839b Update strength image to image gen Pranav Veldurthi 2024-12-16 18:40:36 -05:00
  • e2474c3f15 fail if we never get the desired node count Alex Cheema 2024-12-16 21:59:02 +00:00
  • 1b14be6013 make device_capabilities async running on a thread pool Alex Cheema 2024-12-16 21:17:30 +00:00
  • 036224f877 add topology to tinychat ui Alex Cheema 2024-12-16 21:17:12 +00:00
  • b17faa8199 dont broadcast every single process_tensor Alex Cheema 2024-12-16 20:54:38 +00:00
  • 35d90d947c Merge remote-tracking branch 'origin/main' into runners Alex Cheema 2024-12-16 20:04:03 +00:00
  • 8d94b8ae12 trigger test Alex Cheema 2024-12-16 20:03:22 +00:00
  • 99a70f1045 Merge commit: trigger test Alex Cheema 2024-12-16 20:01:23 +00:00
  • bd0febe35f Merge commit: trigger test Alex Cheema 2024-12-16 20:01:09 +00:00
  • 34ecbbe01c Merge commit: trigger test Alex Cheema 2024-12-16 20:00:50 +00:00
  • 427d0718b3 Merge commit: trigger test Alex Cheema 2024-12-16 20:00:39 +00:00
  • b49c4ca0e5 Merge commit: trigger test Alex Cheema 2024-12-16 20:00:21 +00:00
  • 41eaaec5a9 Merge commit: trigger test Alex Cheema 2024-12-16 20:00:10 +00:00
  • bf1aafdea7 Merge commit: trigger test Alex Cheema 2024-12-16 19:59:51 +00:00
  • bfa06ee9f3 Merge commit: trigger test Alex Cheema 2024-12-16 19:59:39 +00:00
  • c0534b67c3 Merge commit: trigger test Alex Cheema 2024-12-16 19:59:08 +00:00
  • 063964aab3 remove redundant sample_logits, put back opaque status for process_prompt so we have a way of preemptively starting downloads Alex Cheema 2024-12-16 19:33:14 +00:00
  • 804ad4705a upgrade mlx Alex Cheema 2024-12-16 17:52:53 +00:00
  • c9ded9ba96 optimise networking, remove bloat Alex Cheema 2024-12-16 17:52:35 +00:00
  • 64365d684f one two and three m4 pro clusters Alex Cheema 2024-12-15 15:11:47 +00:00
  • 9397464fad add commit to results Alex Cheema 2024-12-15 15:06:24 +00:00
  • 08912d1b64 Only collect topology if peers changed Nel Nibcord 2024-12-15 02:50:34 -08:00
  • 06c2e236b8 rip out stats bloat Alex Cheema 2024-12-14 21:40:14 +00:00
  • cb4615c95d fix SendNewToken Alex Cheema 2024-12-14 21:11:00 +00:00
  • f55a53ae7e one token at a time Alex Cheema 2024-12-14 21:06:41 +00:00
  • 4d9d4ad05a separate line by branch name githubactions Gary 2024-12-15 14:54:06 +00:00
  • cfedcec3a6 Merge pull request #558 from blindcrone/topology-on-change v1.0 blindcrone 2024-12-15 05:00:12 -08:00
  • 470f961fbb Only collect topology if peers changed Nel Nibcord 2024-12-15 02:50:34 -08:00
  • 25b4af70e0 Merge branch 'main' into runners Gary 2024-12-14 20:48:58 +00:00
  • a93092105c set max-generate-tokens to 250 Alex Cheema 2024-12-14 19:10:03 +00:00
  • 0c6ab35333 increase timeout of http request in bench.py up to 10 mins Alex Cheema 2024-12-14 18:33:41 +00:00
  • 149849f94e format metric changes nicely in discord Alex Cheema 2024-12-13 22:36:55 +00:00
  • 0fa8f1f5bb discord notifications on new benchmark runs Alex Cheema 2024-12-13 22:27:11 +00:00
  • 72be5e4bd5 Merge pull request #556 from exo-explore/fixtestmodelhelpers Alex Cheema 2024-12-13 21:29:18 +00:00
  • b0e079b36a fix counts in testmodelhelpers fixtestmodelhelpers Alex Cheema 2024-12-13 17:46:44 +00:00
  • e5d54c77a9 add llama-3.3-70b to 3 M4 Pro cluster Alex Cheema 2024-12-12 18:51:26 +00:00
  • 6016e1185f 100x faster dashboard Alex Cheema 2024-12-12 18:50:34 +00:00
  • 2ff4638122 Merge remote-tracking branch 'origin/main' into runners Alex Cheema 2024-12-12 17:14:40 +00:00