# MiniMax-M3.1-Flash

> A multimodal coding model with a 1M context window and tunable thinking depth, served through RealRelay's OpenAI-compatible endpoint.

## Key facts

- Model author: MiniMax
- Input: $0.15 per 1M tokens
- Output: $0.60 per 1M tokens
- Cached input: $0.015 per 1M tokens
- Context window: 1,000,000 tokens
- Input modalities: image, text, video
- Output modalities: text
- Capabilities: reasoning, tools
- Model IDs: `minimax/MiniMax-M3.1-Flash`
- Model page: https://www.realrelay.ai/model/minimax/minimax-m3.1-flash

## Overview

MiniMax-M3.1-Flash is a multimodal coding model from MiniMax. It accepts image, text and video input and returns text, with a 1,000,000 token context window and tunable thinking depth.

On RealRelay, MiniMax-M3.1-Flash is callable with the model ID minimax/MiniMax-M3.1-Flash through an OpenAI-compatible endpoint, so requests use the standard chat format with the model field set to that ID.

## Capabilities

MiniMax-M3.1-Flash takes image, text and video input and produces text output, so text-only and multimodal prompts use the same model ID.

The model supports reasoning with tunable thinking depth, letting you set how much thinking is applied to a request.

Tool calling is supported, so the model can request tools defined in a request, and prompts of up to 1,000,000 tokens of context can be sent in a single call.

## Pricing

Input is priced at $0.15 per 1M tokens and output at $0.60 per 1M tokens.

Cached input reads are priced at $0.015 per 1M tokens, so a reused prompt prefix is billed at the cache read rate.

## Use cases

Coding requests that pair text with images or video fit the input modalities the model accepts, such as sending a screenshot together with a written task.

Because the context window is 1,000,000 tokens, long files and lengthy conversations can be passed in one request, and thinking depth can be tuned per request for tasks that need more or less reasoning.

## 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-M3.1-Flash", "messages": [{"role": "user", "content": "Hello"}]}'
```

## FAQ

### What is MiniMax-M3.1-Flash?

MiniMax-M3.1-Flash is a multimodal coding model from MiniMax with a 1,000,000 token context window and tunable thinking depth. It accepts image, text and video input and returns text.

### How is MiniMax-M3.1-Flash priced?

Input is $0.15 per 1M tokens and output is $0.60 per 1M tokens. Cached input reads are $0.015 per 1M tokens.

### What is the context window of MiniMax-M3.1-Flash?

The context window is 1,000,000 tokens, and that full window can be sent in one request through RealRelay's OpenAI-compatible endpoint.

### Which inputs and outputs does MiniMax-M3.1-Flash support?

The model accepts image, text and video input and produces text output.

### Does MiniMax-M3.1-Flash support tool calling and reasoning?

Yes. The model supports tool calling and reasoning, and the thinking depth of a request can be tuned.

### How do I call MiniMax-M3.1-Flash through the OpenAI-compatible API?

Send a standard chat request to RealRelay's OpenAI-compatible endpoint with the model field set to minimax/MiniMax-M3.1-Flash.