[AutoRound] Support w8a8 scheme in auto-round and add example

[mengniwang95]

SUMMARY:

• Support w8a8 static fp8 quantization in auto-round modifier
• Support w8a8 dynamic fp8 quantization in auto-round modifier
• Add llama4 w8a8 static/dynamic fp8 quantization example

note: this PR depends on the auto-round PR: intel/auto-round#1161

TEST PLAN:
run the ut: ./tests/llmcompressor/transformers/autoround/test_autoround_oneshot.py

[gemini-code-assist]

Summary of Changes

Hello @mengniwang95, 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 significantly enhances the AutoRoundModifier by introducing support for both static and dynamic w8a8 FP8 quantization schemes. It also includes new Llama4 examples to showcase these capabilities, making the quantization process more versatile for various model architectures and precision requirements.

Highlights

• W8A8 Static FP8 Quantization Support: The AutoRound modifier now supports w8a8 static FP8 quantization, expanding its capabilities for model compression.
• W8A8 Dynamic FP8 Quantization Support: The AutoRound modifier now supports w8a8 dynamic FP8 quantization, offering more flexibility in how models are compressed.
• Llama4 FP8 Quantization Examples: New examples have been added to demonstrate w8a8 static and dynamic FP8 quantization specifically for Llama4 models, providing practical guidance for users.

🧠 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.

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. ↩

[github-actions]

👋 Hi! Thank you for contributing to llm-compressor. Please add the ready label when the PR is ready for review.

Note: This is required to complete the testing suite, please only add the label once the PR is code complete and local testing has been performed.

[gemini-code-assist]

Code Review

This pull request adds support for w8a8 static and dynamic fp8 quantization to the AutoRound modifier, including new examples and tests. The core logic changes in base.py are significant. My review identified a few issues: a copy-paste error in an example file, a couple of bugs in the new activation quantization logic within base.py, and a removed assertion in a test file that weakens test coverage. Addressing these points will improve the correctness and clarity of the implementation.

[xin3he]

How about using act_dynamic = getattr(activation_args, "dynamic", None)?

[mengniwang95]

There are default values in QuantizationArgs for each parameter. If we use getattr, I think all similar codes should be replaced to keep aligned.