MiniMax M3 Thinking

minimax/minimax-m3:thinking

MiniMax M3 Thinking is the adaptive-thinking version of MiniMax's open-weights frontier model for coding, agent workflows, tool use, long-context tasks, and native multimodal understanding. MiniMax reports 59.0% on SWE-Bench Pro and 66.0% on Terminal Bench 2.1, with Sparse Attention designed to scale context to 1M. It starts with a 512K context cap on Hide Router for now.

Vision
Reasoning
Tool calling
Structured output

Chat with this model Create an API key

Pricing

Includes our margin. Charged per request from your balance.

Input: $0.253 per 1M tokens
Output: $1.06 per 1M tokens
Cache read: $0.055 per 1M tokens
Typical message: $0.00082

Details

Provider: MiniMax
Context: 512K tokens
Max output: 80K tokens
Added: Jun 1, 2026

Use it from the API

Replace $HIDE_API_KEY with a key from the Developers page. Base URL https://hide.st/api/v1.

curl https://hide.st/api/v1/chat/completions \
  -H "Authorization: Bearer $HIDE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax/minimax-m3:thinking",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

Pricing

Details

Use it from the API

Search