Kimi K3 is Live on CKimi K3 is Live on Canopy Wave. Start BuildingArrow
MiniMax M3 API
CODECHAT

MiniMax M3 API

All You Need to Know About MiniMax M3 API

Overview

Model Provider:MiniMax
Model Type:Code/Chat
State:Ready

Key Specs

Quantization:BF16
Parameters:427B
Context:1M
Pricing:$0.30 input / $1.20 output / $0.06 cached
Try Model API
Quick Start
Reserve Dedicated Endpoint

Introduction

MiniMax-M3 is a native multimodal model with 1M context. It has ~428B parameters and ~23B activated parameters.

Highlights:

  • Native Multimodality: M3 undergoes mixed-modality training from the very first step, enabling deeper semantic fusion across text, image, and video.
  • Context Scaling via Sparse Attention: M3 introduces MiniMax Sparse Attention (MSA) to improve long context efficiency. M3 delivers 9× prefill and 15× decode speedups compared to M2 at 1M context, reducing per-token compute to 1/20.
  • Coding & Cowork Capability: M3 achieves frontier-level performance across long-horizon agentic benchmarks, excelling in both coding and cowork.

MiniMax M3 API Usage

Model

Endpoint

minimax/minimax-m3


        1
        curl -X POST https://inference.canopywave.io/v1/chat/completions \
      
        2
          -H "Content-Type: application/json" \
      
        3
          -H "Authorization: Bearer $CANOPYWAVE_API_KEY" \
      
        4
          -d '{
      
        5
            "model": "minimax/minimax-m3",
      
        6
            "messages": [
      
        7
              {"role": "user", "content": "tell me a story"}
      
        8
            ],
      
        9
            "max_tokens": 1000,
      
        10
            "temperature": 0.7
      
        11
          }'