# MiniMax-M2.5

> MiniMax's text model optimized for code generation and refactoring, with reasoning and tool calling.

## Key facts

- Model author: MiniMax
- Input: $0.2851 per 1M tokens (starting price)
- Output: $1.1404 per 1M tokens (starting price)
- Cached input: $0.0285 per 1M tokens
- Context window: 204,800 tokens
- Input modalities: text
- Output modalities: text
- Capabilities: reasoning, tools
- Released: 2026-02-12
- Model IDs: `jd/MiniMax-M2.5`, `minimax/MiniMax-M2.5`
- Model page: https://www.realrelay.ai/model/minimax/minimax-m2.5

## Overview

MiniMax-M2.5 is a text model from MiniMax optimized for code generation and refactoring. It supports reasoning and tool calling, takes text input and returns text. It offers a 204,800-token context window, large enough to hold many source files, and it was released on February 12, 2026.

It has a 204,800-token context window. It accepts text input and returns text. It was released on February 12, 2026.

## Capabilities

MiniMax-M2.5 combines reasoning with tool calling for coding work: it can plan an edit, call tools to inspect or run code and use the results to finish the change.

Only text goes in and out, so code, logs and written instructions are the natural inputs. Cached input reads are priced lower than regular input, which helps with repeated repository context.

## Pricing

MiniMax-M2.5 is priced from $0.2851 per 1M input tokens and $1.1404 per 1M output tokens. Cached input reads cost $0.0285 per 1M tokens. The prices shown are starting prices and can vary.

MiniMax-M2.5 is available through RealRelay's OpenAI-compatible API.

## Use cases

Use MiniMax-M2.5 to generate code from a description, fill in tests and write scripts, with the 204,800-token window holding the surrounding files for reference.

It is also a fit for refactoring, where the model reasons about how to restructure a module and calls tools to apply the change and check that it still works.

## Call it

OpenAI-compatible chat completions:

```bash
curl https://www.realrelay.ai/v1/chat/completions \
  -H "Authorization: Bearer $REALRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "minimax/MiniMax-M2.5", "messages": [{"role": "user", "content": "Hello"}]}'
```

## FAQ

### What is MiniMax-M2.5?

MiniMax-M2.5 is a text model from MiniMax optimized for code generation and refactoring. It supports reasoning and tool calling, takes text input and returns text. It offers a 204,800-token context window, large enough to hold many source files, and it was released on February 12, 2026.

### How much does MiniMax-M2.5 cost?

MiniMax-M2.5 is priced from $0.2851 per 1M input tokens and $1.1404 per 1M output tokens. Cached input reads cost $0.0285 per 1M tokens. The prices shown are starting prices and can vary.

### What is the context window of MiniMax-M2.5?

MiniMax-M2.5 has a 204,800-token context window.

### What inputs and outputs does MiniMax-M2.5 support?

MiniMax-M2.5 accepts text input and returns text.

### Does MiniMax-M2.5 support tool calling?

Yes. MiniMax-M2.5 supports tool calling and reasoning, with text input and text output, and is optimized for code generation and refactoring.

### When was MiniMax-M2.5 released?

MiniMax-M2.5 was released on February 12, 2026.

### How do I call MiniMax-M2.5 through the API?

Call MiniMax-M2.5 through RealRelay's OpenAI-compatible API by setting the model field to minimax/MiniMax-M2.5.