[codex] Update MiniMax M3 B200 EAGLE3 FlashInfer image

[cquil11]

Summary

• update minimaxm3-fp8-b200-vllm-mtp to vllm/vllm-openai:minimax-m3-0618
• enable FlashInfer TRT-LLM attention with FP8 indexer KV and KV cache
• use TRITON_ATTN for the EAGLE3 drafter and the image's default FlashInfer TRT-LLM MoE backend

Validation

• benchmark script passes bash -n
• MiniMax M3 B200-DGXC vLLM sweep generation succeeds

---

Note

Medium Risk
Benchmark-only changes, but they alter attention/MoE backends and patch installed vLLM at job start—incorrect flags or patch anchors would break engine init on B200 MTP sweeps.

Overview
Brings minimaxm3-fp8-b200-vllm-mtp in line with the non-MTP B200 0618 vLLM image: CI image tag, benchmark serve flags, and docs/changelog.

The MTP benchmark script now uses FlashInfer TRT-LLM attention with FP8 indexer KV and KV cache, drops explicit Marlin MoE on TP-only runs (image default MoE), and applies the same runtime MSA prefill .contiguous() patch on sparse_attention_msa.py as the base B200 recipe. The EAGLE3 draft head attention backend changes from FLASH_ATTN to TRITON_ATTN so spec decode matches the new target stack.

^{Reviewed by Cursor Bugbot for commit 6e6fdc9. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Thanks for the contribution! For vLLM & SGLang, please ensure that your recipes is similar to the official vLLM recipes and/or the SGLang cookbook

• https://docs.sglang.io/cookbook/intro
• https://recipes.vllm.ai/

If it is not, please create a PR first before we can merge your single node PR into the master branch. Let's ensure that the documentation is first class such that the entire ML community can benefit from your hard work! Thank you

• https://github.com/vllm-project/recipes
• https://github.com/sgl-project/sglang/tree/main/docs_new

PR authors are responsible for ensuring that after merging, all GitHub Action jobs fully pass. A lot of the time, failures are just flakes and simply re-running the failed jobs will fix it. If re-running failed jobs is attempted, PR authors are responsible for ensuring it passes. See GitHub's docs on re-running failed jobs: https://docs.github.com/en/actions/how-tos/manage-workflow-runs/re-run-workflows-and-jobs#re-running-failed-jobs-in-a-workflow

As a rule of thumb, generally, PR authors should request a review & get a PR approval from the respective companies' CODEOWNERS before requesting a review from core maintainers.

If additional help is needed, PR authors can reach out to core maintainers over Slack.

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27792967781
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27792967781

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27796128080
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27796128080

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27805369633
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27805369633

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27805369633
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27805369633

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27805369633
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27805369633

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27805369633
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27805369633

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27805369633
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27805369633

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27805369633
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27805369633

[Oseltamivir]

/reuse-sweep-run

[Oseltamivir]

lgtm