MiniMax: MiniMax M1

Text input Text output
Author's Description

MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference. It leverages a hybrid Mixture-of-Experts (MoE) architecture paired with a custom "lightning attention" mechanism, allowing it to process long sequences—up to 1 million tokens—while maintaining competitive FLOP efficiency. With 456 billion total parameters and 45.9B active per token, this variant is optimized for complex, multi-step reasoning tasks. Trained via a custom reinforcement learning pipeline (CISPO), M1 excels in long-context understanding, software engineering, agentic tool use, and mathematical reasoning. Benchmarks show strong performance across FullStackBench, SWE-bench, MATH, GPQA, and TAU-Bench, often outperforming other open models like DeepSeek R1 and Qwen3-235B.

Key Specifications
Cost
$$$$$
Context
1M
Released
Jun 17, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Reasoning Include Reasoning Top P Max Tokens Temperature
Features

This model supports the following features:

Reasoning
Performance Summary

MiniMax M1 is a large-scale, open-weight reasoning model excelling in complex tasks, particularly those requiring extended context. While its speed ranking places it among the slowest models (2nd percentile) and its pricing is positioned at premium levels (15th percentile), it demonstrates exceptional reliability with a 98% success rate across benchmarks. The model shows strong performance in General Knowledge (99.8% accuracy), Email Classification (100% accuracy), and Reasoning (93.9% accuracy), indicating robust understanding and problem-solving capabilities. It also performs well in Coding (91.0% accuracy) and Mathematics (87.5% accuracy), aligning with its optimization for software engineering and mathematical reasoning. A notable weakness is its 89.8% accuracy in Hallucinations, suggesting room for improvement in acknowledging uncertainty. Despite its longer response times and higher cost, M1's high accuracy across critical reasoning and knowledge domains, coupled with its outstanding reliability, makes it a powerful tool for demanding applications.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.3
Completion $1.65

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Minimax
Minimax | minimax/minimax-m1 1M $0.3 / 1M tokens $1.65 / 1M tokens
Novita
Novita | minimax/minimax-m1 1M $0.55 / 1M tokens $2.2 / 1M tokens
SiliconFlow
SiliconFlow | minimax/minimax-m1 131K $0.55 / 1M tokens $2.2 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by minimax