GPT-6 Sol & Claude Opus 5.5 are liveGPT-6 Sol at half the price of 5.6 Sol

MiniMax M3 API

LLM

MiniMax M3 is MiniMax's long-context flagship, with a 1M window, tool use, and a second rate tier past 512K input.

  • Input · <=512K$0.24 USDOfficial price$0.3 USD
  • Output · <=512K$0.96 USDOfficial price$1.2 USD
  • Input · 512K-1M$0.48 USDOfficial price$0.6 USD
  • Output · 512K-1M$1.92 USDOfficial price$2.4 USD
  • Savings20%
  • Context1M
  • Last verified2026-09-04
  • View tool pricing and billing rules

What are we making today?

Bring an idea, a question, or a first draft.

Enter to send · Shift + Enter for a new line

01

Where it fits

MiniMax M3 is MiniMax's long-context flagship, with a 1M window, tool use, and a second rate tier past 512K input.

Choose it for large-document and long-session agent work, keeping requests under 512K input where the lower tier applies.

02

Model capabilities

Tool use

Can select and call tools in agent workflows that support structured tool use.

03

OpenAI-compatible API

Keep your client and change the base URL, API key, and model ID.

curl
curl https://api.omniakey.com/v1/chat/completions \
  -H "Authorization: Bearer $OMNIAKEY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "MiniMax-M3",
    "messages": [{"role": "user", "content": "Explain this codebase"}],
    "stream": true
  }'

04

Compatible coding tools

Claude Code
Codex
Cursor
Cline
Aider

More MiniMax models

Frequently asked questions

How much does MiniMax M3 cost on OmniaKey?

Rates vary with input context length: <=512K: Input $0.24 USD, Output $0.96 USD, Cache hit / 1M $0.048 USD; 512K-1M: Input $0.48 USD, Output $1.92 USD, Cache hit / 1M $0.096 USD.

What is the context window for MiniMax M3?

The listed context window is 1M. Actual usable limits can also depend on the client and upstream request format.

Which coding tools work with MiniMax M3?

The current catalog lists compatibility with Claude Code, Codex, Cursor, Cline, Aider. Use the model ID shown on this page when the client supports model selection.