Skip to content

feat(minimax): add MiniMax-M3 model#1

Open
kapelame wants to merge 1 commit into
mainfrom
feat/minimax-m3
Open

feat(minimax): add MiniMax-M3 model#1
kapelame wants to merge 1 commit into
mainfrom
feat/minimax-m3

Conversation

@kapelame

@kapelame kapelame commented Jun 1, 2026

Copy link
Copy Markdown
Owner

Summary

Add MiniMax M3 flagship to the minimax model provider plugin.

  • New models/minimax/models/llm/minimax-m3.yaml: 512K context, 128K max output, vision feature (M3 is native multimodal), agent-thought / tool-call / thinking.
  • Pricing: 4.2 / 16.8 RMB per M tokens (input/output) — 2× M2.7, matching the provider's RMB convention.
  • _position.yaml: M3 listed first (flagship).
  • llm.py: _MODEL_ALIASES maps minimax-m3MiniMax-M3 (server-side id).
  • manifest.yaml: version 0.0.20 → 0.0.21 (required by the marketplace version-exists CI check).

Test

  • minimax-m3.yaml parses cleanly; vision feature + 512K context + RMB pricing verified.
  • Endpoint already Anthropic-compatible (api.minimax.io/anthropic) — no transport change.

Internal review PR within my fork.

MiniMax M3 flagship — 512K context, 128K max output, native multimodal
(adds vision feature), reasoning/tool-call. Pricing: 4.2/16.8 RMB per M
(input/output). New minimax-m3.yaml + _position + _MODEL_ALIASES entry +
manifest version bump (0.0.20 -> 0.0.21).
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant