feat/integrate tunix v1

[pathfinder-pf]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @pathfinder-pf, 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 integrates tunix v1 by performing a series of foundational updates across the project. It modernizes dependency management, significantly refactors the scheduler's threading model for improved robustness, and adopts flax.nnx to enhance the structure and state management of various neural network models. Additionally, it improves asynchronous operations, refines sampling parameter handling, and expands test coverage for cache behavior.

Highlights

• Dependency Updates: The pyproject.toml file has been updated to specify newer versions for key dependencies: flax is now required to be >=0.12.0 (previously ~=0.10.7), and jax[tpu] is updated from ~=0.6.2 to a more general jax[tpu].
• Scheduler Refactoring: The scheduler initialization and management logic has been significantly refactored. The run_scheduler_thread function has been replaced by run_scheduler_loop_thread_after_create and scheduler_loop_after_create, streamlining how scheduler threads are launched and their information is retrieved, moving away from pipe-based communication.
• Flax NNX Integration: The codebase now integrates flax.nnx for managing model layers and attention backend metadata. This involves importing flax.nnx, using nnx.data() for dynamic state in FlashAttention, and wrapping model layers in nnx.List() within Llama, QWen, Qwen2, Qwen3, and QWen3Moe models, enhancing module composition and state management.
• Asynchronous Event Loop Handling: The engine.py file now includes robust handling for asynchronous event loops, specifically addressing RuntimeError when asyncio.get_event_loop() is called without an active loop, by creating a new event loop if necessary.
• Sampling Parameter Enhancements: The sampling_batch_info.py file has been updated to correctly handle sampling_seeds during precompilation, including padding with a default seed and incorporating them into the SamplingMetadata.
• New Cache Miss Test Cases: Two new test cases, test_cache_miss_prefill and test_cache_miss_decode, have been added to test_features.py to verify the cache_miss_count in the response metadata during both prefill and decode phases of generation.

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. ↩