Add b200 benchmark numbers to Qwen3-Next.md

[coreylowman]

I also have some numbers for 2 speculative tokens, and for different levels of tensor parallelism on the B200. B200 can go all the way down to tp=1, but its much slower due to small kv cache.

Any requests for numbers to go in here?

[gemini-code-assist]

Summary of Changes

Hello @coreylowman, 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 enhances the documentation for the Qwen3-Next model by incorporating new benchmark data obtained from B200 hardware. The added information provides crucial performance metrics under various configurations, including standard serving and speculative decoding, which will help users understand the model's capabilities and efficiency on cutting-edge accelerators.

Highlights

• B200 Benchmark Results: New benchmark numbers for the Qwen3-Next model running on B200 GPUs have been added, providing performance insights for this advanced hardware.
• Speculative Decoding Benchmarks: The pull request includes specific benchmark results for speculative decoding (MTP) on the B200, showcasing performance with 4 speculative tokens.
• Updated Benchmark Concurrency: An existing benchmark command's --max-concurrency parameter was updated from 10 to 256, reflecting a higher load scenario.

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 in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

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 issue 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 adds valuable benchmark results for the Qwen3-Next model on B200 hardware, both with and without Multi-Token Prediction (MTP). The changes are clear and provide useful performance data. I've made a few minor suggestions to improve consistency and clarity in the documentation.

[vadiklyutiy]

I think it is a little bit early to publish the performance numbers. We still have some serious performance gaps. IMO better to wait a bit when we fix this performance issues

[coreylowman]

I think it is a little bit early to publish the performance numbers. We still have some serious performance gaps. IMO better to wait a bit when we fix this performance issues

Okay! @simon-mo asked for a PR so i'm happy to wait for whatever performance improvements. Was pretty easy to produce (though I can't promise we'll have b200s at any given point)

[simon-mo]

There were some more discussion in the slack thread

[coreylowman]

Ahhh caught up now. should i close this PR?