AR
feat: added minimax provider
archestra-ai/archestra#1938

Overview

This PR introduces the MiniMax (Global/Intl) LLM provider to the Archestra platform. It enables MiniMax as a first-class citizen for both the LLM Proxy and the Chat interface, supporting tool invocation, streaming, and TOON compression.

Changes

1. Backend: LLM Proxy & Adapters

  • Adapter Implementation: Created backend/src/routes/proxy/adapterV2/minimax.ts which implements the LLMRequestAdapter, LLMResponseAdapter, and LLMStreamAdapter following the OpenAI-compatible schema used by MiniMax.
  • Unified Proxy Handler: Registered MiniMax routes in backend/src/routes/proxy/minimax.ts using the generic handleLLMProxy logic.
  • Model Listing: Added fetchMiniMaxModels to backend/src/routes/chat-models.ts with a hardcoded fallback to ensure availability of MiniMax-M2.1 and associated models.
  • Tokenizer Support: Updated backend/src/tokenizers/index.ts to use TiktokenTokenizer for MiniMax.

2. Frontend: UI & Instructions

  • API Key Management: Updated ChatApiKeyForm to support MiniMax, including icons, placeholder formats, and links to the MiniMax international console.
  • Model Selector:
    • Integrated MiniMax models into the ModelSelector grouping.
    • Updated ModelSelectorLogo types and local logo mappings to display the MiniMax branding correctly.
  • Proxy Instructions: Added MiniMax-specific base URL instructions to the ProxyConnectionInstructions component to help users configure their external clients.

3. Interaction & Analytics

  • Interaction Mapping: Updated interaction.utils.ts to route minimax:chatCompletions through the OpenAiChatCompletionInteraction class for consistent metrics and cost calculation.

4. Quality Assurance & Tests

  • Unit Tests: Created backend/src/routes/proxy/adapterV2/minimax.test.ts with 23 test cases covering:
    • Request/Response parsing.
    • Tool call conversion.
    • Usage extraction.
    • Error handling.
  • E2E Verification: Verified the proxy endpoints (/v1/proxy/minimax/...) handle both streaming and non-streaming requests correctly.

Verification Checklist

  • MiniMax models appear in the Chat model selector.
  • MiniMax API keys can be added/saved via Settings.
  • Proxy instructions display the correct URL transformation for MiniMax.
  • Streaming chat works with real/mock MiniMax responses.
  • Unit tests pass: pnpm test minimax.test.ts.

Related Issues

Fixes #1855 /claim #1855

Claim

Total prize pool $50
Total paid $0
Status Pending
Submitted January 08, 2026
Last updated January 08, 2026

Contributors

PR

Pranjal Negi

@Pranjal6955

100%

Sponsors

AR

Archestra

@archestra-ai

$50