Skip to content
This repository was archived by the owner on Mar 29, 2026. It is now read-only.

feat: add MiniMax LLM provider with M2.7 default#1118

Open
octo-patch wants to merge 3 commits into
vanna-ai:mainfrom
octo-patch:feature/add-minimax-provider
Open

feat: add MiniMax LLM provider with M2.7 default#1118
octo-patch wants to merge 3 commits into
vanna-ai:mainfrom
octo-patch:feature/add-minimax-provider

Conversation

@octo-patch

@octo-patch octo-patch commented Mar 14, 2026

Copy link
Copy Markdown

Summary

  • Add MiniMax as an LLM provider integration using OpenAI-compatible API
  • Support MiniMax-M2.7 (default), MiniMax-M2.7-highspeed, MiniMax-M2.5, and MiniMax-M2.5-highspeed models
  • Temperature clamping for MiniMax API compatibility
  • Full streaming and tool-call support

Changes

  • New MinimaxLlmService class implementing LlmService interface
  • Default model set to MiniMax-M2.7 (latest flagship with enhanced reasoning and coding)
  • 15 unit tests covering initialization, payload building, and tool validation
  • Optional dependency via pip install vanna[minimax]

Why

MiniMax-M2.7 is the latest flagship model with enhanced reasoning and coding capabilities. The integration uses the standard OpenAI SDK pointed at MiniMax's compatible endpoint.

Testing

  • All 15 unit tests passing
  • Integration tested with MiniMax API

PR Bot and others added 2 commits March 14, 2026 22:25
Add MiniMax as a new LLM provider for Vanna, enabling users to leverage
MiniMax's models (MiniMax-M2.5, MiniMax-M2.5-highspeed) with 204K context
windows through their OpenAI-compatible API.

Changes:
- New integration module at src/vanna/integrations/minimax/
- Implements LlmService interface with streaming and tool support
- Handles MiniMax-specific temperature constraints (0.0, 1.0]
- Adds minimax optional dependency in pyproject.toml
- Includes comprehensive unit tests (15 tests)
- Updates README to list MiniMax as a supported LLM provider

Co-Authored-By: octo-patch <noreply@github.com>
- Set MiniMax-M2.7 as the new default model (was M2.5)
- Update docstrings to reference M2.7 model family
- Update unit tests to verify M2.7 default
- All previous models remain available as alternatives
@octo-patch octo-patch changed the title Add MiniMax LLM provider integration feat: add MiniMax LLM provider with M2.7 default Mar 18, 2026
- Add MiniMax-M3 to model list and set as default
- Keep MiniMax-M2.7 and MiniMax-M2.7-highspeed
- Remove older models (M2.5/M2.1/M2/M1)
- Update related tests
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant