[docs] Update readme

[Jintao-Huang]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request primarily focuses on updating documentation and example scripts to reflect the latest capabilities and best practices of the ms-swift framework. The changes ensure that users have accurate information regarding supported models, dependency versions, and command-line arguments for training and deployment, enhancing clarity and usability for both English and Chinese speaking audiences.

Highlights

• Documentation Updates: Updated the number of supported multimodal large models from 300+ to 400+ across various README and quick-start guides (English and Chinese versions).
• Dependency Versioning: Adjusted the recommended and required transformers library versions in documentation tables and the install_all.sh script, specifically updating to transformers<5.3.0 in the installation script and transformers>=5.2.0 in model requirements.
• Example Command Updates: Modified the example command-line interface for fine-tuning, including changing the default model from Qwen/Qwen2.5-7B-Instruct to Qwen/Qwen3-4B-Instruct-2507, reducing the estimated memory usage from 22GB to 13GB, and removing the --system argument.
• Megatron Parallelism Description: Refined the description of Megatron Parallelism benefits for MoE models, changing 'acceleration up to 10x' to 'significantly boost MoE model training speed' in the documentation.
• Qwen3 Omni Example: Added the --experts_impl grouped_mm argument to the zero3.sh example script for Qwen3 Omni models.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• README.md
  • Updated the count of supported multimodal large models from 300+ to 400+.
  • Revised the description of Megatron Parallelism for MoE models.
  • Updated the recommended transformers version in the installation table.
  • Modified the example swift sft command to use Qwen/Qwen3-4B-Instruct-2507, adjusted memory estimate to 13GB, and removed the --system argument.
  • Emphasized 'ModelScope' in the default download method description.
• README_CN.md
  • Updated the count of supported multimodal large models from 300+ to 400+.
  • Revised the description of Megatron Parallelism for MoE models.
  • Updated the recommended transformers version in the installation table.
  • Modified the example swift sft command to use Qwen/Qwen3-4B-Instruct-2507, adjusted memory estimate to 13GB, and removed the --system argument.
  • Emphasized 'ModelScope' in the default download method description.
• docs/source/GetStarted/Quick-start.md
  • Updated the count of supported multimodal large models from 300+ to 400+.
  • Revised the description of Megatron Parallelism for MoE models.
  • Modified the example swift sft command to use Qwen/Qwen3-4B-Instruct-2507, adjusted memory estimate to 13GB, and removed the --system argument.
  • Emphasized 'ModelScope' in the default download method description.
• docs/source/GetStarted/SWIFT-installation.md
  • Updated the recommended transformers version in the installation table.
• docs/source/Instruction/Supported-models-and-datasets.md
  • Updated the transformers version requirement for Qwen3.5 MoE models from >=5.2.0.dev to >=5.2.0.
• docs/source/Megatron-SWIFT/Quick-start.md
  • Updated the recommended transformers version in the installation table.
• docs/source_en/GetStarted/Quick-start.md
  • Updated the count of supported multimodal large models from 300+ to 400+.
  • Revised the description of Megatron Parallelism for MoE models.
  • Modified the example swift sft command to use Qwen/Qwen3-4B-Instruct-2507, adjusted memory estimate to 13GB, and removed the --system argument.
  • Emphasized 'ModelScope' in the default download method description.
• docs/source_en/GetStarted/SWIFT-installation.md
  • Updated the recommended transformers version in the installation table.
• docs/source_en/Instruction/Supported-models-and-datasets.md
  • Updated the transformers version requirement for Qwen3.5 MoE models from >=5.2.0.dev to >=5.2.0.
• docs/source_en/Megatron-SWIFT/Quick-start.md
  • Updated the recommended transformers version in the installation table.
• examples/models/qwen3_omni/zero3.sh
  • Added the --experts_impl grouped_mm argument to the example script.
• requirements/install_all.sh
  • Updated the transformers installation constraint to "transformers<5.3.0".
• swift/model/models/qwen.py
  • Updated the transformers version requirement for Qwen3_5MoeLoader from >=5.2.0.dev to >=5.2.0.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request primarily updates documentation across multiple files, including READMEs and quick start guides, to reflect new model support and updated examples. While most changes are straightforward content updates, I've identified several inconsistencies in model names and supported model counts across different documentation files. These should be synchronized to prevent user confusion. I've added specific comments with suggestions on the affected files. Other changes, like pinning dependency versions, are good improvements.

[Jintao-Huang]

/gemini review

[gemini-code-assist]

Code Review

This pull request primarily updates documentation across several files (README.md, README_CN.md, and others in docs/) to reflect the latest supported models, dependency versions, and improved examples. The changes are consistent and enhance the clarity of the documentation. I've found one minor issue in an example script where an argument is duplicated, which I've commented on.

[gemini-code-assist]

The argument --experts_impl grouped_mm is duplicated. It's already specified on line 22. Please remove this redundant line to avoid potential issues and improve script clarity.