ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
906 commits 391 branches 58 tags 304 MiB
Commit graph

906 commits

This branch
This branch
All branches
Author SHA1 Message Date
Shine1i
a2dde15367 merge nightly 2026-03-09 00:32:33 +01:00
Roland Tannous
2c879bf4a4 Merge pull request #320 from unslothai/fix/stale-dataset-split-on-switch 
Fix/stale dataset split on switch
 2026-03-09 00:14:09 +04:00
Roland Tannous
1bf0d39f3d Merge pull request #323 from unslothai/fix/vision-dataset-search-filter 
fixing update model type
 2026-03-09 00:13:50 +04:00
Roland Tannous
d98d4da6c8 Merge pull request #223 from unslothai/feature/support-for-audio-models 
Adding support for audio llms
 2026-03-09 00:09:59 +04:00
Roland Tannous
7b76fccb9b fix: loosen executorch pin for python 3.13 compat 2026-03-08 19:56:09 +00:00
Roland Tannous
a1778d6655 fix: replace is_dataset_multimodal with is_dataset_image/is_dataset_audio in training orchestrator 2026-03-08 19:40:00 +00:00
Roland Tannous
d10012d9fb silence pip check output 2026-03-08 19:22:31 +00:00
Roland Tannous
0f11415c22 make pip check non-fatal for known third-party conflicts 2026-03-08 19:21:28 +00:00
Manan17
80b704d7b7 Audio_VLM bug fix 2026-03-08 19:14:07 +00:00
Roland Tannous
5a52a0131a add extras-no-deps install step for audio model support 2026-03-08 18:49:55 +00:00
Roland Tannous
7ee81dd7df feat: route audio inference (TTS, ASR, Whisper) through orchestrator/worker subprocess 2026-03-08 18:25:27 +00:00
Roland Tannous
f4393ed3e5 fix: pin streamdown package versions to avoid type mismatch 2026-03-08 16:29:28 +00:00
Wasim Yousef Said
efaae9dc4c Merge pull request #308 from unslothai/fix/browser-autofill-hf-token 
Prevent browser credential autofill in HF token fields
 2026-03-08 16:51:11 +01:00
Wasim Yousef Said
2001abfb26 Merge pull request #331 from unslothai/fix/slider-fill-alignment 
Align slider fill bar with thumb across value range
 2026-03-08 16:50:38 +01:00
Roland Tannous
1e39e7d05f fix: handle structured audio part type in chat adapter 2026-03-08 14:18:16 +00:00
Roland Tannous
1435dbaf59 merge nightly into audio branch (mock test) 2026-03-08 10:23:44 +00:00
imagineer99
075bfe961b fix: track live slider values for uncontrolled mode and scope fill to horizontal 2026-03-08 10:06:17 +00:00
Manan17
ef714f010e adding export support 2026-03-08 04:18:20 +00:00
Roland Tannous
ff56a5f785 Merge pull request #318 from unslothai/fix/recharts-dimension-warning 
Fix Recharts -1 dimension warning on chart mount
 2026-03-08 03:34:58 +04:00
Roland Tannous
aab35f2ed3 Merge pull request #324 from unslothai/feature/subprocess-isolation-version-switching 
Subprocess isolation for training, inference, and export with automatic transformers version switching
 2026-03-08 03:34:12 +04:00
Roland Tannous
a7c34b42be fix: clear stale model state on failed inference subprocess reload 2026-03-07 23:32:53 +00:00
Roland Tannous
4f766bbe25 fix: reset checkpoint metadata on failed export checkpoint reload 2026-03-07 23:29:34 +00:00
Roland Tannous
1dab9f57b6 fix: validate pip exit codes for .venv_t5 installs in setup.ps1 2026-03-07 23:25:13 +00:00
Roland Tannous
adf0fa6f81 add .venv_t5/ to .gitignore 2026-03-07 23:21:48 +00:00
Manan17
6487f81113 check fir gated repo 2026-03-07 21:32:50 +00:00
Manan17
905f989521 fixing sesame model 2026-03-07 19:06:00 +00:00
Roland Tannous
8454e6dd2b fix: scope dataloader_num_workers=0 to Windows + transformers 5.x only 2026-03-07 17:55:59 +00:00
Roland Tannous
ef9184c731 fix: prevent training hang on Windows by adding triton-windows support 2026-03-07 17:53:36 +00:00
Roland Tannous
e25705a211 fix: propagate PYTHONPATH to child subprocesses, revert tokenizer patching 2026-03-07 11:28:24 +00:00
Roland Tannous
9330588015 fix: patch TokenizersBackend in export output after save_pretrained 2026-03-07 10:57:51 +00:00
Roland Tannous
76c78afb8f fix: patch TokenizersBackend by model name - Qwen3.5→Qwen2Tokenizer, GLM→PreTrainedTokenizer 2026-03-07 10:29:59 +00:00
Roland Tannous
d60cd2843f fix: patch Qwen3.5 broken tokenizer_class TokenizersBackend across all backends 2026-03-07 09:43:25 +00:00
Roland Tannous
29fa91be07 fix: bump transformers to 5.2.0 and pin huggingface_hub in setup.ps1 2026-03-07 09:12:12 +00:00
Roland Tannous
bd60562145 fix: bump transformers 5.x pin from 5.1.0 to 5.2.0 for Qwen3.5 support 2026-03-07 09:10:09 +00:00
Roland Tannous
0b3397cc3a fix: fail fast if runtime pip install of transformers 5.x fails 2026-03-07 08:40:25 +00:00
Roland Tannous
f3aeceeb24 fix: join prior pump thread before starting new training job 2026-03-07 08:37:03 +00:00
Roland Tannous
f7a3092cbd fix: correct project root depth in model_config.py vision check 2026-03-07 08:15:29 +00:00
Roland Tannous
728420b290 fix: drain stale events from resp_queue after generation cancel 2026-03-07 08:12:16 +00:00
Roland Tannous
25b51fad3b fix: wait for training shutdown before export load, clear stop flag on reset 
1. Export route: stop_training() only signals the subprocess — wait up to
   30s for it to actually exit before loading the export checkpoint, avoiding
   a GPU memory race.

2. Training reset: clear _should_stop so /api/train/status returns phase=idle
   instead of staying stuck on phase=stopped after a user-triggered stop.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 04:16:10 +00:00
Roland Tannous
f45f1f74c0 fix: add /v1 proxy entry to vite dev server config 
Without this, /v1/chat/completions requests in local dev are served by
Vite instead of being proxied to the FastAPI backend.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 04:09:28 +00:00
Roland Tannous
609e3168a1 fix: serialize generation with _gen_lock to prevent concurrent queue readers 
Two overlapping /chat/completions requests could both read from the shared
resp_queue, consuming and dropping each other's token events. Replace the
request_id filtering (which silently dropped non-matching messages) with a
threading.Lock that serializes generation — correct for single-GPU inference.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 04:06:51 +00:00
Roland Tannous
9470957bb9 fix: log final GGUF file locations after relocation 2026-03-06 18:04:27 +00:00
Roland Tannous
5a828ebd43 fix: increase export timeout to 1 hour for large model GGUF conversion 2026-03-06 17:59:42 +00:00
Roland Tannous
4b7ad23b3a feat: broaden Qwen3.5 matching to cover entire family 2026-03-06 16:48:28 +00:00
Roland Tannous
ed1e63c814 feat: add Qwen3.5-35B-A3B and Qwen3-Next to transformers 5.x model list 2026-03-06 10:54:48 +00:00
Manan17
be517cc958 derive effective model type from isVisionModel for dataset search filterin 2026-03-06 08:47:56 +00:00
Shine1i
5e5feb5c00 feat(recipe-studio, validators): tweak OXC validator with lint suppression support and improve error normalization logic 2026-03-06 09:40:53 +01:00
Roland Tannous
d910759121 feat: add OpenAI-compatible /v1/chat/completions endpoint 2026-03-06 07:48:09 +00:00
Manan17
cb3f3f4d0c fixing update model type 2026-03-06 07:44:57 +00:00
imagineer99
15efb0f235 fix: harden chart container sizing with legacy event rechecks 2026-03-06 07:16:36 +00:00