80.2% SWE-Bench Verified

MiniMax-M2.5 API

State-of-the-art open-source coding model. 128K context, FP8 precision, served on 8x H100 GPUs. OpenAI-compatible API.

curl https://api.minimax.villamarket.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-m2.5",
    "messages": [{"role": "user", "content": "Write a Python quicksort"}],
    "stream": true
  }'

Install the CLI

Chat, code, and run skills from your terminal with mm

curl -fsSL minimax.villamarket.ai/install | sh

Why MiniMax-M2.5?

Top Coding Performance

80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench. Best-in-class for code generation and understanding.

128K Context Window

Full 128K token context with FP8 precision on 8x H100 GPUs. Handle entire codebases in a single request.

OpenAI Compatible

Drop-in replacement for OpenAI API. Works with Claude Code, Cursor, Aider, Continue, and any OpenAI-compatible tool.

Tool Calling & Reasoning

Native function calling with structured outputs. Think blocks show the model's reasoning process.

Open Source Model

Built on MiniMaxAI/MiniMax-M2.5, a fully open-source MoE model. No vendor lock-in.

Usage-Based Pricing

$0.30/M input tokens, $1.20/M output tokens. Pay only for what you use within your budget.

Pricing

Simple, transparent pricing. Start free, scale as you grow.

Free

$0/mo

$5/month API budget
5 requests per minute
5 API keys
128K context
Community support

Pro

$20/mo

$20/month API budget
16 requests per minute
5 API keys
128K context
Priority support
Usage analytics

Enterprise

$100/mo

$100/month API budget
Unlimited requests per minute
Unlimited API keys
128K context
Dedicated support
Usage analytics
Custom integrations

Token pricing: $0.30/M input, $1.20/M output. Budget limits prevent unexpected charges.

Ready to start building?

Get your API key in seconds. No credit card required for the free tier.

Create Free Account