Skip to content

[megatron] support bailing_v25 megatron #9442

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

Merged
Jintao-Huang merged 4 commits into from
May 28, 2026
Merged

[megatron] support bailing_v25 megatron #9442

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
75dd946
support bailing_v25 megatron
Jintao-Huang May 28, 2026
2bda596
update
Jintao-Huang May 28, 2026
e4db2e4
update
Jintao-Huang May 28, 2026
08b8edf
lint pass
Jintao-Huang May 28, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 8 additions & 5 deletions docs/source/Instruction/Supported-models-and-datasets.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -353,6 +353,9 @@
|[ZhipuAI/LongWriter-llama3.1-8b](https://modelscope.cn/models/ZhipuAI/LongWriter-llama3.1-8b)|llama|longwriter_llama|transformers>=4.43|✔|-|[zai-org/LongWriter-llama3.1-8b](https://huggingface.co/zai-org/LongWriter-llama3.1-8b)|
|[deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://modelscope.cn/models/deepseek-ai/DeepSeek-R1-Distill-Llama-8B)|llama|deepseek_r1|-|✔|-|[deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B)|
|[deepseek-ai/DeepSeek-R1-Distill-Llama-70B](https://modelscope.cn/models/deepseek-ai/DeepSeek-R1-Distill-Llama-70B)|llama|deepseek_r1|-|✔|-|[deepseek-ai/DeepSeek-R1-Distill-Llama-70B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B)|
|[OpenBMB/MiniCPM5-1B](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B)|
|[OpenBMB/MiniCPM5-1B-Base](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-Base)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B-Base](https://huggingface.co/openbmb/MiniCPM5-1B-Base)|
|[OpenBMB/MiniCPM5-1B-SFT](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-SFT)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B-SFT](https://huggingface.co/openbmb/MiniCPM5-1B-SFT)|
Comment thread
Jintao-Huang marked this conversation as resolved.
Comment on lines +356 to +358

@gemini-code-assist gemini-code-assist Bot May 28, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The Hugging Face transformers library does not currently have a version 5.6 (the latest is in the 4.x range). This is likely a typo for transformers>=4.40 or similar. Please verify and correct the required version.

Suggested change
|[OpenBMB/MiniCPM5-1B](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B)|
|[OpenBMB/MiniCPM5-1B-Base](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-Base)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B-Base](https://huggingface.co/openbmb/MiniCPM5-1B-Base)|
|[OpenBMB/MiniCPM5-1B-SFT](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-SFT)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B-SFT](https://huggingface.co/openbmb/MiniCPM5-1B-SFT)|
|[OpenBMB/MiniCPM5-1B](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B)|llama|minicpm5|transformers>=4.40|✔|-|[openbmb/MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B)|
|[OpenBMB/MiniCPM5-1B-Base](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-Base)|llama|minicpm5|transformers>=4.40|✔|-|[openbmb/MiniCPM5-1B-Base](https://huggingface.co/openbmb/MiniCPM5-1B-Base)|
|[OpenBMB/MiniCPM5-1B-SFT](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-SFT)|llama|minicpm5|transformers>=4.40|✔|-|[openbmb/MiniCPM5-1B-SFT](https://huggingface.co/openbmb/MiniCPM5-1B-SFT)|

|[LLM-Research/Reflection-Llama-3.1-70B](https://modelscope.cn/models/LLM-Research/Reflection-Llama-3.1-70B)|llama|reflection|transformers>=4.43|✔|-|[mattshumer/Reflection-Llama-3.1-70B](https://huggingface.co/mattshumer/Reflection-Llama-3.1-70B)|
|[01ai/Yi-6B](https://modelscope.cn/models/01ai/Yi-6B)|yi|chatml|-|✔|-|[01-ai/Yi-6B](https://huggingface.co/01-ai/Yi-6B)|
|[01ai/Yi-6B-200K](https://modelscope.cn/models/01ai/Yi-6B-200K)|yi|chatml|-|✔|-|[01-ai/Yi-6B-200K](https://huggingface.co/01-ai/Yi-6B-200K)|
Expand Down Expand Up @@ -608,11 +611,11 @@
|[inclusionAI/Ling-mini-base-2.0](https://modelscope.cn/models/inclusionAI/Ling-mini-base-2.0)|bailing_moe|ling2|-|✔|-|[inclusionAI/Ling-mini-base-2.0](https://huggingface.co/inclusionAI/Ling-mini-base-2.0)|
|[inclusionAI/Ling-1T](https://modelscope.cn/models/inclusionAI/Ling-1T)|bailing_moe|ling2|-|✔|-|[inclusionAI/Ling-1T](https://huggingface.co/inclusionAI/Ling-1T)|
|[inclusionAI/Ring-mini-2.0](https://modelscope.cn/models/inclusionAI/Ring-mini-2.0)|bailing_moe|ring2|-|✔|-|[inclusionAI/Ring-mini-2.0](https://huggingface.co/inclusionAI/Ring-mini-2.0)|
|[inclusionAI/Ling-2.5-1T](https://modelscope.cn/models/inclusionAI/Ling-2.5-1T)|bailing_hybrid|ling2|-|✘|-|[inclusionAI/Ling-2.5-1T](https://huggingface.co/inclusionAI/Ling-2.5-1T)|
|[inclusionAI/Ling-2.6-1T](https://modelscope.cn/models/inclusionAI/Ling-2.6-1T)|bailing_hybrid|ling2|-|✘|-|[inclusionAI/Ling-2.6-1T](https://huggingface.co/inclusionAI/Ling-2.6-1T)|
|[inclusionAI/Ling-2.6-flash](https://modelscope.cn/models/inclusionAI/Ling-2.6-flash)|bailing_hybrid|ling2|-|✘|-|[inclusionAI/Ling-2.6-flash](https://huggingface.co/inclusionAI/Ling-2.6-flash)|
|[inclusionAI/Ring-2.5-1T](https://modelscope.cn/models/inclusionAI/Ring-2.5-1T)|bailing_hybrid|ring2_5|-|✘|-|[inclusionAI/Ring-2.5-1T](https://huggingface.co/inclusionAI/Ring-2.5-1T)|
|[inclusionAI/Ring-2.6-1T](https://modelscope.cn/models/inclusionAI/Ring-2.6-1T)|bailing_hybrid|ring2_5|-|✘|-|[inclusionAI/Ring-2.6-1T](https://huggingface.co/inclusionAI/Ring-2.6-1T)|
|[inclusionAI/Ling-2.5-1T](https://modelscope.cn/models/inclusionAI/Ling-2.5-1T)|bailing_hybrid|ling2|-|✔|-|[inclusionAI/Ling-2.5-1T](https://huggingface.co/inclusionAI/Ling-2.5-1T)|
|[inclusionAI/Ling-2.6-1T](https://modelscope.cn/models/inclusionAI/Ling-2.6-1T)|bailing_hybrid|ling2|-|✔|-|[inclusionAI/Ling-2.6-1T](https://huggingface.co/inclusionAI/Ling-2.6-1T)|
|[inclusionAI/Ling-2.6-flash](https://modelscope.cn/models/inclusionAI/Ling-2.6-flash)|bailing_hybrid|ling2|-|✔|-|[inclusionAI/Ling-2.6-flash](https://huggingface.co/inclusionAI/Ling-2.6-flash)|
|[inclusionAI/Ring-2.5-1T](https://modelscope.cn/models/inclusionAI/Ring-2.5-1T)|bailing_hybrid|ring2_5|-|✔|-|[inclusionAI/Ring-2.5-1T](https://huggingface.co/inclusionAI/Ring-2.5-1T)|
|[inclusionAI/Ring-2.6-1T](https://modelscope.cn/models/inclusionAI/Ring-2.6-1T)|bailing_hybrid|ring2_5|-|✔|-|[inclusionAI/Ring-2.6-1T](https://huggingface.co/inclusionAI/Ring-2.6-1T)|
|[IEITYuan/Yuan2.0-2B-hf](https://modelscope.cn/models/IEITYuan/Yuan2.0-2B-hf)|yuan2|yuan|-|✘|-|[IEITYuan/Yuan2-2B-hf](https://huggingface.co/IEITYuan/Yuan2-2B-hf)|
|[IEITYuan/Yuan2.0-51B-hf](https://modelscope.cn/models/IEITYuan/Yuan2.0-51B-hf)|yuan2|yuan|-|✘|-|[IEITYuan/Yuan2-51B-hf](https://huggingface.co/IEITYuan/Yuan2-51B-hf)|
|[IEITYuan/Yuan2.0-102B-hf](https://modelscope.cn/models/IEITYuan/Yuan2.0-102B-hf)|yuan2|yuan|-|✘|-|[IEITYuan/Yuan2-102B-hf](https://huggingface.co/IEITYuan/Yuan2-102B-hf)|
Expand Down
13 changes: 8 additions & 5 deletions docs/source_en/Instruction/Supported-models-and-datasets.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -354,6 +354,9 @@ The table below introduces the models integrated with ms-swift:
|[ZhipuAI/LongWriter-llama3.1-8b](https://modelscope.cn/models/ZhipuAI/LongWriter-llama3.1-8b)|llama|longwriter_llama|transformers>=4.43|✔|-|[zai-org/LongWriter-llama3.1-8b](https://huggingface.co/zai-org/LongWriter-llama3.1-8b)|
|[deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://modelscope.cn/models/deepseek-ai/DeepSeek-R1-Distill-Llama-8B)|llama|deepseek_r1|-|✔|-|[deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B)|
|[deepseek-ai/DeepSeek-R1-Distill-Llama-70B](https://modelscope.cn/models/deepseek-ai/DeepSeek-R1-Distill-Llama-70B)|llama|deepseek_r1|-|✔|-|[deepseek-ai/DeepSeek-R1-Distill-Llama-70B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B)|
|[OpenBMB/MiniCPM5-1B](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B)|
|[OpenBMB/MiniCPM5-1B-Base](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-Base)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B-Base](https://huggingface.co/openbmb/MiniCPM5-1B-Base)|
|[OpenBMB/MiniCPM5-1B-SFT](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-SFT)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B-SFT](https://huggingface.co/openbmb/MiniCPM5-1B-SFT)|
Comment thread
Jintao-Huang marked this conversation as resolved.
Comment on lines +357 to +359

@gemini-code-assist gemini-code-assist Bot May 28, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The Hugging Face transformers library does not currently have a version 5.6 (the latest is in the 4.x range). This is likely a typo for transformers>=4.40 or similar. Please verify and correct the required version.

Suggested change
|[OpenBMB/MiniCPM5-1B](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B)|
|[OpenBMB/MiniCPM5-1B-Base](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-Base)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B-Base](https://huggingface.co/openbmb/MiniCPM5-1B-Base)|
|[OpenBMB/MiniCPM5-1B-SFT](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-SFT)|llama|minicpm5|transformers>=5.6|✔|-|[openbmb/MiniCPM5-1B-SFT](https://huggingface.co/openbmb/MiniCPM5-1B-SFT)|
|[OpenBMB/MiniCPM5-1B](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B)|llama|minicpm5|transformers>=4.40|✔|-|[openbmb/MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B)|
|[OpenBMB/MiniCPM5-1B-Base](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-Base)|llama|minicpm5|transformers>=4.40|✔|-|[openbmb/MiniCPM5-1B-Base](https://huggingface.co/openbmb/MiniCPM5-1B-Base)|
|[OpenBMB/MiniCPM5-1B-SFT](https://modelscope.cn/models/OpenBMB/MiniCPM5-1B-SFT)|llama|minicpm5|transformers>=4.40|✔|-|[openbmb/MiniCPM5-1B-SFT](https://huggingface.co/openbmb/MiniCPM5-1B-SFT)|

|[LLM-Research/Reflection-Llama-3.1-70B](https://modelscope.cn/models/LLM-Research/Reflection-Llama-3.1-70B)|llama|reflection|transformers>=4.43|✔|-|[mattshumer/Reflection-Llama-3.1-70B](https://huggingface.co/mattshumer/Reflection-Llama-3.1-70B)|
|[01ai/Yi-6B](https://modelscope.cn/models/01ai/Yi-6B)|yi|chatml|-|✔|-|[01-ai/Yi-6B](https://huggingface.co/01-ai/Yi-6B)|
|[01ai/Yi-6B-200K](https://modelscope.cn/models/01ai/Yi-6B-200K)|yi|chatml|-|✔|-|[01-ai/Yi-6B-200K](https://huggingface.co/01-ai/Yi-6B-200K)|
Expand Down Expand Up @@ -609,11 +612,11 @@ The table below introduces the models integrated with ms-swift:
|[inclusionAI/Ling-mini-base-2.0](https://modelscope.cn/models/inclusionAI/Ling-mini-base-2.0)|bailing_moe|ling2|-|✔|-|[inclusionAI/Ling-mini-base-2.0](https://huggingface.co/inclusionAI/Ling-mini-base-2.0)|
|[inclusionAI/Ling-1T](https://modelscope.cn/models/inclusionAI/Ling-1T)|bailing_moe|ling2|-|✔|-|[inclusionAI/Ling-1T](https://huggingface.co/inclusionAI/Ling-1T)|
|[inclusionAI/Ring-mini-2.0](https://modelscope.cn/models/inclusionAI/Ring-mini-2.0)|bailing_moe|ring2|-|✔|-|[inclusionAI/Ring-mini-2.0](https://huggingface.co/inclusionAI/Ring-mini-2.0)|
|[inclusionAI/Ling-2.5-1T](https://modelscope.cn/models/inclusionAI/Ling-2.5-1T)|bailing_hybrid|ling2|-|✘|-|[inclusionAI/Ling-2.5-1T](https://huggingface.co/inclusionAI/Ling-2.5-1T)|
|[inclusionAI/Ling-2.6-1T](https://modelscope.cn/models/inclusionAI/Ling-2.6-1T)|bailing_hybrid|ling2|-|✘|-|[inclusionAI/Ling-2.6-1T](https://huggingface.co/inclusionAI/Ling-2.6-1T)|
|[inclusionAI/Ling-2.6-flash](https://modelscope.cn/models/inclusionAI/Ling-2.6-flash)|bailing_hybrid|ling2|-|✘|-|[inclusionAI/Ling-2.6-flash](https://huggingface.co/inclusionAI/Ling-2.6-flash)|
|[inclusionAI/Ring-2.5-1T](https://modelscope.cn/models/inclusionAI/Ring-2.5-1T)|bailing_hybrid|ring2_5|-|✘|-|[inclusionAI/Ring-2.5-1T](https://huggingface.co/inclusionAI/Ring-2.5-1T)|
|[inclusionAI/Ring-2.6-1T](https://modelscope.cn/models/inclusionAI/Ring-2.6-1T)|bailing_hybrid|ring2_5|-|✘|-|[inclusionAI/Ring-2.6-1T](https://huggingface.co/inclusionAI/Ring-2.6-1T)|
|[inclusionAI/Ling-2.5-1T](https://modelscope.cn/models/inclusionAI/Ling-2.5-1T)|bailing_hybrid|ling2|-|✔|-|[inclusionAI/Ling-2.5-1T](https://huggingface.co/inclusionAI/Ling-2.5-1T)|
|[inclusionAI/Ling-2.6-1T](https://modelscope.cn/models/inclusionAI/Ling-2.6-1T)|bailing_hybrid|ling2|-|✔|-|[inclusionAI/Ling-2.6-1T](https://huggingface.co/inclusionAI/Ling-2.6-1T)|
|[inclusionAI/Ling-2.6-flash](https://modelscope.cn/models/inclusionAI/Ling-2.6-flash)|bailing_hybrid|ling2|-|✔|-|[inclusionAI/Ling-2.6-flash](https://huggingface.co/inclusionAI/Ling-2.6-flash)|
|[inclusionAI/Ring-2.5-1T](https://modelscope.cn/models/inclusionAI/Ring-2.5-1T)|bailing_hybrid|ring2_5|-|✔|-|[inclusionAI/Ring-2.5-1T](https://huggingface.co/inclusionAI/Ring-2.5-1T)|
|[inclusionAI/Ring-2.6-1T](https://modelscope.cn/models/inclusionAI/Ring-2.6-1T)|bailing_hybrid|ring2_5|-|✔|-|[inclusionAI/Ring-2.6-1T](https://huggingface.co/inclusionAI/Ring-2.6-1T)|
|[IEITYuan/Yuan2.0-2B-hf](https://modelscope.cn/models/IEITYuan/Yuan2.0-2B-hf)|yuan2|yuan|-|✘|-|[IEITYuan/Yuan2-2B-hf](https://huggingface.co/IEITYuan/Yuan2-2B-hf)|
|[IEITYuan/Yuan2.0-51B-hf](https://modelscope.cn/models/IEITYuan/Yuan2.0-51B-hf)|yuan2|yuan|-|✘|-|[IEITYuan/Yuan2-51B-hf](https://huggingface.co/IEITYuan/Yuan2-51B-hf)|
|[IEITYuan/Yuan2.0-102B-hf](https://modelscope.cn/models/IEITYuan/Yuan2.0-102B-hf)|yuan2|yuan|-|✘|-|[IEITYuan/Yuan2-102B-hf](https://huggingface.co/IEITYuan/Yuan2-102B-hf)|
Expand Down
6 changes: 4 additions & 2 deletions swift/megatron/utils/convert_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -230,8 +230,10 @@ def test_convert_precision(args, hf_model, mg_model, template, test_convert_dtyp
hf_model.to('cpu')

template.use_megatron = True
inputs = template.encode(get_examples(test_mm_type), return_length=True)
mg_inputs = to_device(template.data_collator([inputs], padding_to=get_padding_to(args)), 'cuda')
inputs = [
template.encode(get_examples(test_mm_type), return_length=True) for _ in range(2 if args.padding_free else 1)
]
mg_inputs = to_device(template.data_collator(inputs, padding_to=get_padding_to(args)), 'cuda')
mg_model.eval()
# thd
text_position_ids = mg_inputs.pop('text_position_ids', None)
Expand Down
Loading