Skip to content

feat: add MiniMax regional metadata - #994

Open
octo-patch wants to merge 4 commits into
codexu:devfrom
octo-patch:feature/add-minimax-provider
Open

feat: add MiniMax regional metadata#994
octo-patch wants to merge 4 commits into
codexu:devfrom
octo-patch:feature/add-minimax-provider

Conversation

@octo-patch

@octo-patch octo-patch commented Mar 17, 2026

Copy link
Copy Markdown

Reason: The MiniMax preset lacks regional endpoints and current model capability metadata.

Summary

  • add selectable global and China endpoints for both supported API protocols
  • add context, pricing, thinking, and input-modality metadata for the current models
  • extend the focused provider configuration tests

Checks

  • node --test src/app/core/setting/config.spec.mjs
  • ./node_modules/.bin/tsc --noEmit

octo-patch and others added 3 commits March 17, 2026 18:44
- Add MiniMax to baseAiConfig with API base URL and platform link
- MiniMax models (M2.5, M2.5-highspeed) are available via OpenAI-compatible API
- Add unit tests verifying config entry structure
- Add integration tests for chat completion and streaming
- Update integration tests to use MiniMax-M2.7 as primary model
- Add MiniMax-M2.7-highspeed test
- Add backward compatibility test for M2.5
- Keep all previous models as alternatives
- Add MiniMax-M3 to the MiniMax provider models list as the default
- Keep MiniMax-M2.7 and MiniMax-M2.7-highspeed as alternatives
- Remove MiniMax-M2.5 backward compatibility test
- Add unit test verifying M3 is first in the models list

Co-Authored-By: Octopus <liyuan851277048@icloud.com>
@octo-patch octo-patch changed the title feat: add MiniMax as built-in AI provider feat: upgrade MiniMax default model to M3 Jun 2, 2026
@octo-patch octo-patch changed the title feat: upgrade MiniMax default model to M3 feat: add MiniMax regional metadata Aug 12, 2026
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