Skip to content

Pull requests: modelscope/ms-swift

New pull request New
83 Open 3,642 Closed
83 Open 3,642 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

[bugfix] avoid duplicate resize for Qwen3-Omni
#9729 opened Jul 10, 2026 by taking-lying-flat Loading…
[model] support TeleChat3
#9727 opened Jul 10, 2026 by allplepine Loading…
1 of 4 tasks
9
Call destroy_process_group() at the end of the trainer pipelines
#9726 opened Jul 10, 2026 by gusario Contributor Loading…
1 of 4 tasks
4
support optional megatron-bridge backend
#9725 opened Jul 10, 2026 by hjh0119 Collaborator Loading…
11
Support dsa cp
#9720 opened Jul 10, 2026 by Jintao-Huang Collaborator Loading…
1
feat: add R-SWA training support for UnlimitedOCR model
#9714 opened Jul 9, 2026 by z0o0ey Collaborator Loading…
3
fix: flush asynchronous JSONL logs on training end
#9710 opened Jul 9, 2026 by Alwin4Zhang Loading…
4
fix(tuners/lora): re-apply lora_dtype after peft _cast_adapter_dtype (#9694)
#9699 opened Jul 7, 2026 by Anai-Guo Loading…
1
2
Add language-model-only export option
#9670 opened Jun 30, 2026 by prefersung Loading…
2
Fix: MiniCPM-V 4.6 training hangs on text-only samples with DeepSpeed
#9639 opened Jun 24, 2026 by randydl Contributor Loading…
6
[bugfix] guard _repair_ms_bench against an empty messages list
#9608 opened Jun 20, 2026 by he-yufeng Contributor Loading…
[bugfix] guard load_audio against bytes input
#9606 opened Jun 20, 2026 by he-yufeng Contributor Loading…
1 of 4 tasks
4
[bugfix] include image mode and size in tmp image cache key
#9605 opened Jun 20, 2026 by he-yufeng Contributor Loading…
1 of 4 tasks
1
2
Update load mcore checkpoint
#9595 opened Jun 18, 2026 by Jintao-Huang Collaborator Loading…
2
Update gkd_loss.py teacher_prompt supports two formats:
#9562 opened Jun 15, 2026 by yhy19 Loading…
1 of 4 tasks
4
[bugfix] add optional MEDIA_DECODE_TIMEOUT to prevent silent media-decode hang
#9541 opened Jun 11, 2026 by HaozheZhang6 Contributor Loading…
1 of 4 tasks
1
2
[WIP] support tps
#9513 opened Jun 8, 2026 by Jintao-Huang Collaborator Loading…
6
[WIP] refactor trainer
#9510 opened Jun 8, 2026 by Jintao-Huang Collaborator Loading…
14
fix(megatron): pre-initialize NCCL communicator for MoE expert DP group to prevent lazy-init deadlock
#9486 opened Jun 3, 2026 by zb2313 Loading…
3
Add lightweight TGS and MFU logging for training
#9465 opened Jun 1, 2026 by WendaDeng Loading…
3 tasks
9
[compat] Support Qwen2-Audio with newer transformers
#9453 opened May 30, 2026 by MWXGOD Loading…
2 of 10 tasks
7
feat: Add native support for PyTorch Profiler in ms-swift
#9449 opened May 29, 2026 by qq1243196045 Contributor Loading…
1 of 4 tasks
37
[bugfix] Gemma4 suffix: drop trailing newline
#9440 opened May 28, 2026 by shanhaoli Loading…
2 tasks
4
optimize streaming
#9425 opened May 27, 2026 by Jintao-Huang Collaborator Loading…
16
fix: 更新 MindSpeed import 路径 (lite.ops.triton → ops.triton，适配 core_r0.16.0)
#9415 opened May 26, 2026 by Weizhena Loading…
2
ProTip! Exclude everything labeled bug with -label:bug.