MiniMax Provider

vercel-minimax-ai-provider is a community provider that provides access to the latest MiniMax-M2 model from MiniMax.

API keys can be obtained from the MiniMax Platform.

Setup

The MiniMax provider is available via the vercel-minimax-ai-provider module. You can install it with:

pnpm add vercel-minimax-ai-provider

Provider Instance

The MiniMax provider supports two API compatibility modes:

Anthropic-Compatible API (Default)

You can import the default provider instance minimax from vercel-minimax-ai-provider:

import { minimax } from 'vercel-minimax-ai-provider';

Or explicitly use the Anthropic-compatible instance:

import { minimaxAnthropic } from 'vercel-minimax-ai-provider';

Or OpenAI-compatible API format:

import { minimaxOpenAI } from 'vercel-minimax-ai-provider';

Custom Configuration

For custom configuration, you can use the createMinimax (Anthropic-compatible) or createMinimaxOpenAI (OpenAI-compatible) functions:

Anthropic-Compatible Configuration (Default)

import { createMinimax } from 'vercel-minimax-ai-provider';

const minimax = createMinimax({
  apiKey: process.env.MINIMAX_API_KEY,
});

Configuration Options

You can use the following optional settings to customize the MiniMax provider instance:

baseURL string

Use a different URL prefix for API calls.
- Anthropic-compatible default: https://api.minimax.io/anthropic/v1
- OpenAI-compatible default: https://api.minimax.io/v1
apiKey string

API key that is being sent using the Authorization header. It defaults to the MINIMAX_API_KEY environment variable.
headers Record<string,string>

Custom headers to include in the requests.
fetch (input: RequestInfo, init?: RequestInit) => Promise<Response>

Custom fetch implementation.

API Compatibility

MiniMax provides two API formats. Both are included in this package.

The main difference is the API request/response format:

Anthropic format (default): Uses Anthropic Messages API format with anthropic-version header
OpenAI format: Uses standard OpenAI chat completion format

Both formats access the same MiniMax models with the same capabilities.

Model Capabilities

Model	Text Generation	Object Generation	Image Input	Tool Usage	Tool Streaming
`MiniMax-M2`
`MiniMax-M2-Stable`

Please see the MiniMax docs for a full list of available models and their capabilities. The provider accepts any model ID as a string for forward compatibility.

Example Usage

Basic Text Generation

import { minimax } from 'vercel-minimax-ai-provider';
import { generateText } from 'ai';

const result = await generateText({
  model: minimax('MiniMax-M2'),
  prompt: 'Explain quantum computing in simple terms.',
});

console.log(result.text);

Streaming

import { minimax } from 'vercel-minimax-ai-provider';
import { streamText } from 'ai';

const result = streamText({
  model: minimax('MiniMax-M2'),
  prompt: 'Write a short story about a robot learning to paint.',
});

for await (const chunk of result.textStream) {
  console.log(chunk);
}