Minimax M2
Minimax M 2 is a chat model developed by Minimax, exceling at code generation and reasoning tasks, particularly in coding and agentic workflows. Its notable technical trait is a compact, fast, and cost-effective MoE architecture with 230 billion total parameters and 10 billion active parameters.
Input
Output
Context
197K
Max Output
8K
Parameters
228.7B
Technical Specifications
Model TypeChat
Context Window196,608 tokens
Max Output Tokens8,192 tokens
Parameters228.7B
Release DateOct 22, 2025
Training CutoffJun 1, 2024
Licenseother
Open SourceYes
Input Modalities
Text
Output Modalities
Text
Capabilities
Benchmarks
Artificial Analysis
12.5%
HLE
77.7%
GPQA
36.1%
SciCode
82.0%
MMLU Pro
78.3
Math
29.2
Coding
82.6%
LiveCodeBench
28.3
Intelligence
127
Speed (tok/s)
16.9
TTFA (s)
1.1
TTFT (s)
Estimated GPU Requirements for Minimax M2
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Jun 24, 2026
Start building with the right model.
Automatically route workloads to the right model for every task, every time.