feat: add MiniMax as a direct LLM provider

[octo-patch]

Summary

Add MiniMax as a direct LLM provider alongside the AI Gateway, giving users access to MiniMax-M2.5 and MiniMax-M2.5-highspeed models with a 204K context window.

Changes

• lib/ai/models.ts: Add MiniMax-M2.5 and MiniMax-M2.5-highspeed to the curated model list
• lib/ai/providers.ts: Route MiniMax models directly via @ai-sdk/openai (OpenAI-compatible API) instead of through the AI Gateway
• components/multimodal-input.tsx: Add "MiniMax" to provider display names in the model selector UI
• package.json: Add @ai-sdk/openai dependency for direct provider access
• .env.example: Add MINIMAX_API_KEY environment variable
• README.md: Document MiniMax provider support
• tests/: Add MiniMax integration tests and update model selector test regex

How It Works

MiniMax models bypass the AI Gateway and connect directly to the MiniMax API (https://api.minimax.io/v1) using the OpenAI-compatible interface. Users set the MINIMAX_API_KEY environment variable to enable MiniMax models.

// providers.ts - MiniMax models route directly, not via gateway
if (modelId.startsWith("minimax/")) {
  const minimax = getMiniMaxProvider();
  return minimax(modelId.replace("minimax/", ""));
}

API Documentation

• MiniMax OpenAI-Compatible API

[vercel]

@octo-patch is attempting to deploy a commit to the Zeb Hermann's projects Team on Vercel.

A member of the Team first needs to authorize it.

[socket-security]

Review the following changes in direct dependencies. Learn more about Socket for GitHub.

Diff	Package	Supply Chain Security	Vulnerability	Quality	Maintenance	License
	@​ai-sdk/​openai@​3.0.41

View full report