# MiniMax M3

> MSA sparse attention. SWE-Bench Pro 59%, Terminal-Bench 66%. Agentic coding, 1M context, multimodal input.

Human version: https://kymaapi.com/models/minimax-m3
Live JSON: `GET https://kymaapi.com/v1/models` (no auth required)

## Facts

- **Model ID**: `minimax-m3` — pass this as `model` in the request body
- **Creator**: MiniMax
- **Released**: 2026-05-31
- **Context window**: 1M tokens
- **Max output**: 33K tokens per response — a hard ceiling, not a default
- **Price**: $0.3852 in / $1.54 out per 1M, cached input 10%
- **Capabilities**: tools, vision, reasoning, caching

## Call it

```bash
curl https://kymaapi.com/v1/chat/completions \
  -H "Authorization: Bearer $KYMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "minimax-m3", "messages": [{"role": "user", "content": "Hello"}]}'
```

## Positioning

MiniMax M3 is a frontier model optimized for agentic coding workflows and long-context multimodal analysis. Developers building autonomous agents or debugging complex repositories should use it when they need reliable tool execution across a million-token window.

## About

MiniMax M3 uses MSA sparse attention to process text, image, and video inputs within a 1,048,576-token context window. It supports tool calling, structured outputs, and explicit reasoning traces, making it suitable for multi-step coding tasks and repository-level analysis.

On Kyma, the model runs through an OpenAI-compatible endpoint with automatic request failover and exact cost reporting in the usage.cost field. Prompt caching applies a 90% discount to repeated prefixes, and new accounts receive a $0.50 credit to test throughput.

Output generation caps at 32,768 tokens, and median throughput sits around 2 tokens per second. It is not designed for high-speed conversational streaming or low-latency real-time applications.

## Use cases

- **Agentic Code Generation** — Handles multi-step repository edits and tool chaining across long context windows.
- **Multimodal Input Analysis** — Processes text, image, and video inputs to extract structured data or debug visual workflows.
- **Long-Horizon Debugging** — Maintains state over extended sessions to trace errors across large codebases.
- **Structured Output Parsing** — Returns deterministic JSON or XML formats for downstream pipeline integration.

## Not ideal for

Avoid this model for low-latency chat or high-throughput streaming where sub-second response times are required.

## Pick something else when

- You need faster token generation for interactive chat → `gemini-3.5-flash-lite`
- You need lower cost for high-volume text processing → `deepseek-v3`
- You need longer output sequences for document generation → `qwen-3.7-max`

## See also

- All models: https://kymaapi.com/models.md
- Pricing: https://kymaapi.com/pricing.md
- Other models by MiniMax: https://kymaapi.com/models?q=MiniMax