MiniMax-M3

Name in the API

MiniMaxAI/MiniMax-M3

Base model

MiniMaxAI/MiniMax-M3

Weights we run

nvidia/MiniMax-M3-NVFP4 (NVFP4 quantization)

Type

Chat

Input → output

text, image → text

Context length

524,288 tokens

Availability

HPC

Hardware

4 × NVIDIA H100 (94 GB), one node of the Capella HPC cluster

Thinking

Default: on. Our gateway sets chat_template_kwargs: {"thinking_mode": "enabled"}.

reasoning_effort: every value is accepted and ignored. All levels think about the same amount, and even reasoning_effort: "none" does not turn thinking off.

Numeric effort: no.

Turn off: chat_template_kwargs: {"thinking_mode": "disabled"}. chat_template_kwargs: {"enable_thinking": false} has no effect.

Tool calling

Yes

Limits

At most 4 images per request. Video input is not supported.

Example

Chat request, question about an image. Set model to the name above.