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

627 commits

This branch
This branch
All branches
Author SHA1 Message Date
Roland Tannous
2ce63f09c4 Add gguf to toLoraSummary inline type 2026-02-26 10:59:39 +04:00
Roland Tannous
c21cf2ffcf Add GGUF tag for exported models in chat page selector 2026-02-25 19:01:47 +04:00
Roland Tannous
bfb1403032 Relocate GGUF exports into exports/ directory 2026-02-25 18:54:39 +04:00
Roland Tannous
a8b5b7ed58 Fix GGUF models missing from chat page model search 
GGUF was in the global EXCLUDED_TAGS set which filtered it from all
consumers of useHfModelSearch, including the chat page. Move GGUF
exclusion to an opt-in excludeGguf option so only training and
onboarding pages filter out GGUF models.
 2026-02-25 16:21:08 +04:00
Roland Tannous
f92e4a3e1b Merge pull request #261 from unslothai/feat/gguf-llama-cpp-inference 
Add GGUF model inference via llama-server with quantization variant selection
 2026-02-25 16:07:39 +04:00
Roland Tannous
01082b84e5 Merge branch 'nightly' into feat/gguf-llama-cpp-inference 2026-02-25 16:06:03 +04:00
Roland Tannous
a1e064b1c4 Remove UNSLOTH_ENABLE_LOGGING from export pipeline 2026-02-25 16:00:24 +04:00
Roland Tannous
a7fe8a388c Filter GGUF models from training page model selectors 
GGUF models can't be fine-tuned, so hide them from the training/studio
page while keeping them available for inference on the chat page.

- Add "gguf" to EXCLUDED_TAGS in HF model search hook
- Filter local models with .gguf extension or -GGUF in ID
 2026-02-25 15:47:45 +04:00
Roland Tannous
299ce77467 added vision.py patch for vision processor from PR#260 2026-02-25 11:39:17 +00:00
Roland Tannous
cfaa2f2074 Merge pull request #249 from unslothai/fix/section-card-corner-bleed 
Fix: Clip section card overflow to prevent background bleed
 2026-02-25 15:27:46 +04:00
Roland Tannous
cb3e4f2c26 Merge pull request #259 from unslothai/feat/dataset-subsets-split 
Feat/dataset subsets split
 2026-02-25 15:27:12 +04:00
Roland Tannous
96217b5056 Merge pull request #246 from unslothai/fix/dataset-custom-mapping-heuristic 
adding custom mapping according to the chat templates
 2026-02-25 15:26:36 +04:00
Roland Tannous
d2fe02ff04 Merge pull request #260 from unslothai/fix/fix-vision-processor-unsloth-bug 
fix: correct vision.py patch path to unsloth/models/vision.py + add V…
 2026-02-25 15:21:01 +04:00
Shine1i
db11f1a601 style(studio): align card heights and restore dataset advanced section placement 2026-02-25 12:11:06 +01:00
Roland Tannous
40719c4a6f fix: correct vision.py patch path to unsloth/models/vision.py + add VLM processor diagnostic 2026-02-25 11:06:22 +00:00
Roland Tannous
6f0b7bc38a fix: use raw github URL for vision.py patch + add VLM processor diagnostic logging 2026-02-25 10:29:05 +00:00
Shine1i
cb57d48e7e chore: add tour label next to navbar tour icon 2026-02-25 11:23:43 +01:00
Manan17
6e8e70c987 fixing the chatml None error 2026-02-25 10:23:13 +00:00
Shine1i
122311a6b1 fix recipe output path, remove tracked root datasets 2026-02-25 11:19:10 +01:00
Wasim Yousef Said
c67da8f349 Merge pull request #257 from unslothai/feature/chat-model-switch-warning 
feat: chat model switching toast and add image detection logic
 2026-02-25 01:58:56 -08:00
Shine1i
44d6abb36c feat: chat model switching toast and add image detection logic 2026-02-25 10:55:53 +01:00
Wasim Yousef Said
a793660960 Merge pull request #254 from unslothai/feature/theme-fix 
feat: fix markdown rendering, UI adjustments
 2026-02-25 00:48:09 -08:00
Shine1i
8732f3befb feat: fix markdown rendering, UI adjustments 2026-02-25 09:46:13 +01:00
Wasim Yousef Said
468ec99489 Merge pull request #253 from unslothai/feature/theme-fix 
feat: fix dark mode support and refine UI assets
 2026-02-25 00:27:43 -08:00
Shine1i
adcbf78553 feat: fix dark mode support and refine UI assets 2026-02-25 09:23:05 +01:00
Manan17
47fc79df6d My changes for dataset 2026-02-25 08:15:44 +00:00
Manan17
60912e45e6 adding custom mapping according to the chat templates 2026-02-25 07:56:30 +00:00
imagineer99
0b47ab1eab fix: clip section card overflow to prevent background bleed at rounded corners 2026-02-25 03:54:40 +00:00
Roland Tannous
875c6c8094 Merge pull request #247 from unslothai/fix/path-traversal-vuln 
Fix Content-Length crash and path traversal vulnerability in frontend serving
 2026-02-25 05:16:00 +04:00
Roland Tannous
a6f1153f9a fix: replace FileResponse with Response for index.html to prevent Content-Length mismatch and add path traversal guard 2026-02-25 01:05:04 +00:00
Roland Tannous
da8e58fb8f Merge pull request #30 from unslothai/feature/canvas-lab 
Draft: Data Recipes graph editor WIP
 2026-02-25 04:01:53 +04:00
Shine1i
773f7945a6 chore: squircle! tooltip 2026-02-25 00:55:35 +01:00
Shine1i
faa36e9abb feat: update icons and enhance dark mode styling for navbar and recipes 
- Replaced `CookBookIcon` with `ChefHatIcon` in navbar for improved clarity.
- Added dark mode-specific gradient styles to recipe cards for better visual differentiation.
 2026-02-25 00:39:49 +01:00
Shine1i
164560c6c9 feat: improve dark mode styling and simplify navbar 2026-02-25 00:31:20 +01:00
Roland Tannous
7adb69581e Fix GGUF export cwd confusion: remove os.chdir, use absolute paths 
Remove os.chdir(save_directory) from export.py which was causing all of
unsloth-zoo's relative-path internals (check_llama_cpp, use_local_gguf,
_download_convert_hf_to_gguf) to resolve against the export directory
instead of the repo root. This caused llama.cpp to be cloned inside each
export dir and destroyed the repo root's llama-server build on cleanup.

Now passes absolute paths to save_pretrained_gguf so unsloth resolves
llama.cpp from the repo root where setup.sh already built it.

Also builds llama-quantize in setup.sh (needed by unsloth-zoo's export
pipeline) and symlinks it to llama.cpp root for check_llama_cpp().
 2026-02-25 03:30:54 +04:00
Shine1i
2548720c01 Merge branch 'feature/canvas-lab' of https://github.com/unslothai/new-ui-prototype into feature/canvas-lab 
# Conflicts:
#	studio/frontend/bun.lock
 2026-02-25 00:19:19 +01:00
Shine1i
929c7f86e4 feat: add animated theme toggler and refine dark mode styling 2026-02-25 00:18:20 +01:00
Manan17
fdbc60de77 adding custom mapping according to the chat templates 2026-02-24 21:15:56 +00:00
Roland Tannous
0e7c8a2e5e Switch GGUF backend from /v1/completions to /v1/chat/completions 
Fixes two bugs:
1. Chat template tags (<|im_start|>, <|im_end|>) leaking into output
   because /v1/completions treated them as literal text
2. Image hallucination because image_b64 was never passed to llama-server

Now llama-server handles chat templates natively and receives images
as OpenAI-format multimodal content parts for vision models.
 2026-02-24 19:21:01 +04:00
Roland Tannous
ef1cd3ac98 Use llama-server -hf mode, add GGUF variant selector, fix vision detection 
Replace Python-side GGUF download with llama-server's native -hf flag for
HuggingFace repos. Add frontend variant picker so users can choose
quantization (Q4_K_M, Q8_0, BF16, etc.) with file sizes. Fix vision
detection via mmproj files instead of hardcoding is_vision=False.
 2026-02-24 19:03:06 +04:00
Roland Tannous
08aeeaee4b Fix llama-server: build in-tree, fix path resolution, add LD_LIBRARY_PATH 2026-02-24 18:19:29 +04:00
Roland Tannous
4e88092452 Preflight llama-server check before downloading remote GGUF files 2026-02-24 18:02:43 +04:00
Roland Tannous
a900eb9ad7 Fix GGUF detection for HuggingFace repo IDs (not just local paths) 2026-02-24 17:49:09 +04:00
Roland Tannous
70c912d788 Fix CUDA detection for llama-server build on multi-GPU machines 2026-02-24 17:45:17 +04:00
Roland Tannous
a40ebb1aab Add GGUF model inference via llama-server backend 2026-02-24 17:40:05 +04:00
Roland Tannous
3f34996288 Merge branch 'nightly' into feature/canvas-lab 2026-02-24 10:08:13 +00:00
Roland Tannous
7bbb1f0a0b Merge pull request #218 from unslothai/fix/stop-startup-modal 
Added cancel training button on the overlay
 2026-02-24 14:03:25 +04:00
Roland Tannous
d38656139d Merge pull request #241 from unslothai/feature/adding-exported-models-for-chat 
Adding exported model for chat
 2026-02-24 13:55:45 +04:00
Roland Tannous
3ffbee3586 fix(chat): strip /suffix from lora display name and show type tag instead of base model 2026-02-24 09:54:34 +00:00
Roland Tannous
2149bc74ee Merge pull request #232 from unslothai/fix/disable-eval-by-default 
# fix/disable eval by default
 2026-02-24 13:35:11 +04:00