MiniMax M2.7
MiniMax M2.7 is a large language model developed by MiniMax, released on March 18, 2026. It is designed for autonomous productivity and features advanced agentic capabilities, including reasoning and tool-calling support.
MiniMax M2.7 specs and pricing
- Provider: MiniMax
- Released: March 18, 2026
- Context window: 197K tokens
- Max output: 177K tokens
- Input: text
- Output: text
- Price: $0.21 input / $0.84 output per 1M tokens
- Reasoning: yes
- Tool calling: yes
What MiniMax M2.7 is good at
- Complex reasoning tasks
- Automated agentic workflows
- Processing large volumes of text data
To-dos to try with MiniMax M2.7
- Analyze a long document and extract key action items for a project plan
- Use tool-calling to research and summarize information from multiple sources
- Execute a multi-step reasoning process to solve a technical problem
FAQ
- What is the primary focus of the MiniMax M2.7 model? The model is designed for autonomous, real-world productivity and integrates advanced agentic capabilities.
- Does MiniMax M2.7 support tool-calling? Yes, the model is built with native support for tool-calling and reasoning.
- How much does MiniMax M2.7 cost? MiniMax M2.7 costs $0.21 per million input tokens and $0.84 per million output tokens through OpenRouter. In todo.is a typical agent step costs about 0.71 credits ($1 = 100 credits).
- What is the context window of MiniMax M2.7? MiniMax M2.7 can read up to 197K tokens at once and write up to 177K tokens in one reply.
- Can I use MiniMax M2.7 without coding? Yes. In todo.is you pick MiniMax M2.7 as your agent's model and write to-dos in plain words, on the web, the Android app, WhatsApp or Telegram. No API key needed: credits are included, or you can bring your own key.
JavaScript is required to use the todo.is app.