Skip to content

Pull requests: unslothai/unsloth

New pull request New
80 Open 712 Closed
80 Open 712 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][FIX] Vllm guided decoding params
#3662 opened Dec 1, 2025 by Datta0 Loading…
3
Fix: Add support for TRL native dataset formats
#3660 opened Dec 1, 2025 by ved1beta Loading…
1
4
Fix: Pass gradient_checkpointing parameter to model.for_training()
#3659 opened Dec 1, 2025 by sbhavani Loading…
3 tasks
1
3
Fix: Support past_key_values in model.generate for multi-turn conversations
#3653 opened Nov 29, 2025 by vivekkalyanarangan30 Loading…
4 tasks done
1
6
Chunk Across Batch and Context length for logprob calculations for grpo
#3628 opened Nov 21, 2025 by pluesclues Loading…
24
[intel] skip xpu fbgemm fp8
#3625 opened Nov 21, 2025 by leizhenyuan Loading…
3
Feature/raw text dataprep
#3612 opened Nov 18, 2025 by Vangmay Loading…
1
16
fix: unsloth fixes for gfx1151
#3588 opened Nov 12, 2025 by 0xrushi Loading…
1
26
SFT sample packing
#3566 opened Nov 7, 2025 by djsaunde Loading…
@djsaunde
47
Filelocks
#3469 opened Oct 17, 2025 by mmathew23 • Draft
Add Qwen2.5 Coder model support to registry and chat templates
#3436 opened Oct 11, 2025 by Samama-Intellixcore Loading…
Fix/add gpt oss bf16 to mapper
#3425 opened Oct 9, 2025 by rolandtannous Loading…
Added new reference docs - batch 1
#3341 opened Sep 19, 2025 by onel Loading…
1
[ROCm] add rocm dockerfile
#3324 opened Sep 15, 2025 by billishyahao Loading…
3
ROCM support
#3279 opened Sep 5, 2025 by electron271 Loading…
1
21
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
Support for Seq2Seq Models (T5, T5Gemma, etc.)
#3153 opened Aug 14, 2025 by maxzuo • Draft
2
3
Phi‑2 support: partial RoPE, deterministic dropout, loader dispatch, and smoke test
#3125 opened Aug 9, 2025 by MagellaX Loading…
7
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
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
ProTip! Adding no:label will show everything without a label.