fix(minimax): MiniMax M3 model support#3434
Merged
amitksingh1490 merged 5 commits intoJun 2, 2026
Merged
Conversation
github-actions
Bot
added
type: feature
There was a problem hiding this comment.
Pull request overview
Note
Copilot was unable to run its full agentic suite in this review.
Adds support for MiniMax M3 and M3 Pro models, including provider entries and parameter transformer logic with associated tests.
Changes:
- Registers two new MiniMax M3 model entries in the provider catalog.
- Extends
SetMinimaxParamstransformer to detectminimax-m3and apply temperature/top_p/top_k defaults. - Adds three unit tests covering M3, case-insensitivity, and M3 Pro.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
| crates/forge_repo/src/provider/provider.json | Adds minimax.minimax-m3 and minimax.minimax-m3-pro model definitions. |
| crates/forge_app/src/dto/openai/transformers/minimax.rs | Updates transformer to match M3 variants, sets top_k=40, and adds tests. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
- Extended transformer to match minimax-m3 pattern - Added M3 models (minimax-m3, minimax-m3-pro) to provider.json - Added top_k: 40 for M3 models (same as M2.1) - Added comprehensive tests for M3 variants Co-Authored-By: ForgeCode <noreply@forgecode.dev>
ImBIOS
force-pushed
the
feat/add-minimax-m3-support
branch
from
f8e084d to
6178b86
Compare
Co-Authored-By: ForgeCode <noreply@forgecode.dev>
amitksingh1490
changed the title
amitksingh1490
added
type: fix
- Context length: 1M (not 204K) - M3 uses MSA sparse attention - Input modalities: text + image (natively multimodal) - Description: frontier coding, 1M context, MSA, multimodality - Added M3 to minimax direct provider (MiniMax-M3) and Bedrock - Same transformer params as M2.7 (temp=1.0, top_p=0.95, top_k=20) - Added bedrock provider ID test Co-Authored-By: ForgeCode <noreply@forgecode.dev>
The bedrock provider already had minimax-m3 entry at line 1447. Removed the duplicate with incorrect specs (context=204800 instead of 1M, modalities=text instead of text+image). Co-Authored-By: ForgeCode <noreply@forgecode.dev>
Contributor
Author
|
@amitksingh1490, all good now, sir? |
amitksingh1490
approved these changes
Jun 2, 2026
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
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
Add MiniMax M3 model support to forgecode, extending transformer logic and provider configuration to include M3 variants.
Context
Currently, forgecode supports MiniMax M2 series (MiniMax-M2.7) models but lacks M3 support. This feature request was created as issue #3433 after confirming there were no existing issues or PRs for MiniMax M3.
Changes
SetMinimaxParamstransformer to matchminimax-m3pattern in addition to existing M2 patternsprovider.json:minimax.minimax-m3- Flagship model with 204800 context lengthKey Implementation Details
The transformer now checks for both
minimax-m2andminimax-m3patterns. M3 models use the same parameter optimization as M2.1 (top_k: 40) rather than the base M2 settings (top_k: 20).Use Cases
minimax-m3models for enhanced reasoning tasksTesting
Links