MiniMax M2.5 is LiveMiniMax M2.5 is Live on Canopy Wave. Try it Now!DeepSeek V3.1
MiniMax M2 API
CODE

MiniMax M2 API

All You Need To Know About MiniMax M2 API

Overview

Model Provider:MiniMax
Model Type:CODE
State:Take offline

Key Specs

Quantization:FP8
Parameters:230B (10B activated)
Context:128k
Take offline
Quick Start
Reserve Dedicated Endpoint

Introduction

MiniMax-M2 is a compact, high-efficiency Mixture-of-Experts (MoE) model optimized for elite performance in coding and agentic workflows.

With only 10 billion active parameters (out of 230B total), it delivers near-frontier intelligence in areas like end-to-end tool use, multi-step task execution, and complex code generation, matching the performance of much larger models. Its small activation footprint enables fast inference, low latency, and cost-effective deployment, making it ideal for large-scale agents and responsive, reasoning-driven developer assistants.

MiniMax M2 API Usage

Model

Endpoint

minimax/minimax-m2


        1
        curl -X POST https://inference.canopywave.io/v1 \
      
        2
          -H "Content-Type: application/json" \
      
        3
          -H "Authorization: Bearer $CANOPYWAVE_API_KEY" \
      
        4
          -d '{
      
        5
            "model": "minimax/minimax-m2",
      
        6
            "messages": [
      
        7
              {"role": "system", "content": "You are a helpful assistant."},
      
        8
              {"role": "user", "content": "please tell me a story."}
      
        9
            ],
      
        10
          }'
PromotionContact us