[codex] Add MiniMax M3 float32 matmul precision env#553
Draft
jasonlizhengjian wants to merge 1 commit into
Draft
[codex] Add MiniMax M3 float32 matmul precision env#553jasonlizhengjian wants to merge 1 commit into
jasonlizhengjian wants to merge 1 commit into
Conversation
Contributor
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request updates the models/MiniMaxAI/MiniMax-M3.yaml configuration. It updates the date_updated metadata and sets the VLLM_FLOAT32_MATMUL_PRECISION environment variable to high for Hopper hardware overrides and the corresponding guide commands. There are no review comments, so I have no feedback to provide.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds
VLLM_FLOAT32_MATMUL_PRECISION=highto the MiniMax-M3 Hopper recipe path, mirroring the MiniMax-M2.5 H200 setting introduced inb3f013c(Update MiniMax M2.5 H200 recipe (#474)).Changes
extra_envforVLLM_FLOAT32_MATMUL_PRECISION: "high".date_updatedto 2026-06-15.Validation
git diff --checknode scripts/build-recipes-api.mjsVLLM_FLOAT32_MATMUL_PRECISION=high.