Skip to content

Issues: modelscope/ms-swift

Beta
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

763 Open 1,690 Closed
763 Open 1,690 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
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
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

DPO Sequence_parallel_size == 8 Error NotImplementedError
#4420 opened May 29, 2025 by leileilin
支持reinfoce++ enhancement New feature or request
#4419 opened May 29, 2025 by qingchunlizhi
@hjh0119
GRPO训练报错：AssertionError: Forward context is not set. Please use set_forward_context to set the forward context.
#4418 opened May 29, 2025 by jdy18
3
对InternVL3-38B进行GRPO训练，lora和全参两种方式分别需要多少显存？
#4416 opened May 29, 2025 by fourierer
支持多轮DPO吗
#4415 opened May 29, 2025 by leileilin
1
Megatron-SWIFT 的依赖 apex 难装，频繁遇到问题
#4414 opened May 29, 2025 by aqi567
多模态GRPO避免视觉信息重复读取 enhancement New feature or request
#4413 opened May 29, 2025 by davidluciolu
Swift支持在NPU启动序列并行吗
#4412 opened May 29, 2025 by xyzhu1120
Error occurred when saving checkpoints during Qwen3 multi-GPU SFT
#4411 opened May 29, 2025 by yuchengyue
训练完reward model，如何用rm预测单个样本的分数呢？
#4410 opened May 29, 2025 by qq941134965
Saving best ckpt leads to an empty symlink. bug Something isn't working
#4408 opened May 29, 2025 by Chord-Chen-30
通过swift部署的模型随着推理次数增加，显存不断增加
#4407 opened May 29, 2025 by ai-sy41
1
GRPO OOM USE resume_from_checkpoint
#4406 opened May 29, 2025 by warriorandroyal
windows下面web-ui转换的cmd命令无法执行
#4403 opened May 29, 2025 by junshan2009
Unsloth problem with LoRA
#4400 opened May 28, 2025 by hienhayho
megatron train_iters and max_epochs enhancement New feature or request
#4396 opened May 28, 2025 by LaoWangGB
全量微调grpo 相同数量的样本ms-swift效果比unsloth效果差很多 needs more info Additional information or clarification is required to proceed
#4393 opened May 28, 2025 by sxm7078
3
Qwen-Omni 全量微调grpo报错ValueError: max_new_tokens must be greater than 0, but is -16384
#4392 opened May 28, 2025 by TUDelftHao
6
AssertionError: output/v31-20250527-181405/checkpoint-9900 is not an adapter, please try using --model to pass it.
#4389 opened May 28, 2025 by AlexJJJChen
3
qwen3-1..7B GRPO 训练几个轮次后爆显存
#4388 opened May 27, 2025 by yuyaxiong
2
GRPO batchsize=16训练程序 卡在计算ref_per_token_logps
#4387 opened May 27, 2025 by davidluciolu
VLLM Engine Batch 推理咨询
#4386 opened May 27, 2025 by BITDDD
3
seq_cls训练时候开启flash_attn指标大幅度低于不开flash_attn
#4384 opened May 27, 2025 by zhuchenxi
qwen3训练时报错 KeyError: 'ignore_empty_think'
#4382 opened May 27, 2025 by SJLMax
2
reward model的训练能否再详细整理一下文档 documentation Improvements or additions to documentation enhancement New feature or request
#4379 opened May 27, 2025 by 2043078895
2
ProTip! Type g i on any issue or pull request to go back to the issue listing page.