Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Pull requests: unslothai/unsloth-zoo

New pull request New
30 Open 225 Closed
30 Open 225 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[WIP] Mistral 3 Fast Inference
#289 opened Sep 17, 2025 by Datta0 • Draft
fix Gemma3Attention.forward patch fail on transformers versions > 4.55.4
#281 opened Sep 15, 2025 by rolandtannous Loading…
Add video processing utils to vision_utils
#279 opened Sep 13, 2025 by mmathew23 Loading…
Grpo nan fix
#265 opened Sep 5, 2025 by pluesclues Loading…
12
[Intel] remove triton windows for intel
#243 opened Aug 15, 2025 by leizhenyuan Loading…
1
feat: Added Video inference feature into unsloth_zoo/vision_utils.py from qwen-vl-utils
#240 opened Aug 13, 2025 by autinn Loading…
20
Fix: update from disable_adapter to disable_adapters
#221 opened Jul 30, 2025 by wa008 Loading…
6
fix vram estimation when loading a 16bit model with load_in_4bit=True
#193 opened Jul 8, 2025 by yizhe-skyfall Loading…
1
vlm grpo UnslothEfficientGRPO
#188 opened Jul 2, 2025 by GAD-cell Loading…
Change UnslothEfficientGRPO to accept log probs instead of embeddings
#172 opened Jun 19, 2025 by zkpranav Loading…
24
Debug: num_items_in_batch on a different device from loss.
#147 opened May 23, 2025 by KimRass Loading…
2
Convert n_items type from torch.Tensor to int
#139 opened May 15, 2025 by MoonRainy21 Loading…
14
Fix LoRA hot swapping
#116 opened Apr 17, 2025 by pbarker Loading…
2
fix double bos
#106 opened Mar 28, 2025 by tamewild Loading…
2
num_proc=os.cpu_count() when counting unused tokens
#104 opened Mar 27, 2025 by IsaacBreen Loading…
5
saving utils check both model.base_model and base_model.model
#95 opened Mar 22, 2025 by toranb Loading…
VLM Data Collator - Make text & image mixing work efficiently
#89 opened Mar 21, 2025 by mmathew23 Loading…
5
Save compile cache to only home directory
#80 opened Mar 16, 2025 by Datta0 Loading…
2
Add support for multiple instruction patterns in train_on_responses_only
#75 opened Mar 14, 2025 by marcelodiaz558 Loading…
1
Fix Protobuf<4.0 is deprecated, relax dependency
#68 opened Mar 10, 2025 by hoh Loading…
4
Fix VLM train_on_response_only
#60 opened Mar 1, 2025 by coding-famer Loading…
1
37
Fixed getting supported LoRA modules from vLLM.
#58 opened Feb 27, 2025 by LagPixelLOL Loading…
fixed UnslothVisionDataCollator
#56 opened Feb 22, 2025 by KareemMusleh Loading…
update: Update VLMs Data Collator
#55 opened Feb 21, 2025 by rs545837 Loading…
1
Update vision_utils.py
#54 opened Feb 21, 2025 by NinoRisteski Loading…
ProTip! Mix and match filters to narrow down what you’re looking for.