Skip to content

Pull requests: unslothai/unsloth

New pull request New
83 Open 609 Closed
83 Open 609 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

Add support for modules_to_save in FastModel.get_peft_model
#3317 opened Sep 14, 2025 by l1ghtsource Loading…
fix: inference problem on cu124-torch260
#3316 opened Sep 12, 2025 by chengang Loading…
Fix QAT + LoRA fast path, add tests
#3307 opened Sep 10, 2025 by andrewor14 Loading…
ROCM support
#3279 opened Sep 5, 2025 by electron271 Loading…
1
17
Fix grpo nan
#3278 opened Sep 5, 2025 by pluesclues Loading…
5
Fixed duplicate tokenizer_name parameter in FastVisionModel.from_pretrained
#3276 opened Sep 5, 2025 by rjarun8235 Loading…
1
2
feat: removing hard limits on learning rate
#3241 opened Aug 31, 2025 by ysjprojects Loading…
refactor(trainer): add warning for ignored eval_steps
#3234 opened Aug 29, 2025 by MengAiDev Loading…
4
Pass attention mask when doing GRPO
#3176 opened Aug 17, 2025 by Datta0 Loading…
[intel] change windows to remove windows-triton for intel xpu
#3168 opened Aug 15, 2025 by leizhenyuan Loading…
4
Support for Seq2Seq Models (T5, T5Gemma, etc.)
#3153 opened Aug 14, 2025 by maxzuo • Draft
2
2
TRL Updated version of VLM GRPO update along with GSPO
#3132 opened Aug 12, 2025 by pluesclues Loading…
5
Phi‑2 support: partial RoPE, deterministic dropout, loader dispatch, and smoke test
#3125 opened Aug 9, 2025 by MagellaX Loading…
7
update from "mode.disable_adapter()" to "model.disable_adapters()"
#3068 opened Jul 30, 2025 by wa008 Loading…
1
Fast Inference with vLLM for VLMs
#2975 opened Jul 15, 2025 by Datta0 Loading…
19
fix(issue 2950): properly handle spaces in file paths when invoking commands
#2951 opened Jul 13, 2025 by detjonmataj Loading…
1
1
Fix llama.cpp quantize location and execution on Windows.
#2894 opened Jul 7, 2025 by simpolism Loading…
1
5
[Intel] Enable Intel GPU with QLoRA support
#2840 opened Jun 30, 2025 by leizhenyuan • Draft
Online DPO new changes
#2794 opened Jun 23, 2025 by pluesclues Loading…
8
tests for gemma3 patching PR
#2785 opened Jun 22, 2025 by rolandtannous Loading…
2
Avoid materializing the entire logit matrix for logp calculations.
#2772 opened Jun 19, 2025 by zkpranav Loading…
17
Fix beam search for Llama models by adding reorder_cache method
#2753 opened Jun 17, 2025 by amrothemich Loading…
5
[Feature] VLMs support for GRPO
#2752 opened Jun 17, 2025 by GAD-cell Loading…
50
[Feature] Support Sequence Classification
#2704 opened Jun 8, 2025 by rabintiwari45 Loading…
14
Documentation added
#2691 opened Jun 5, 2025 by Aktsvigun Loading…
213
ProTip! Follow long discussions with comments:>50.