Minimax 2.5
MinimaxBalanced
ThinkingTool UseStructured Output
About this model
MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1...
Performance Tier
Balanced
Minimax 2.5 is a balanced model from Minimax : strong performance at a reasonable price.
Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.
Pricing
This model is included in Elosia plansAffordable
Low cost. Suitable for sustained use and high-volume interactions.
| Type | per 1M tokens |
|---|---|
| Input (prompt) | $0.150 |
| Output (completion) | $1.15 |
Capabilities
Context Length197K
Max Output Tokens197K
TokenizerOther
Inputtext
Outputtext
Release DateFebruary 12, 2026
Benchmarks
General Intelligence
MMLU
88%
GPQA Diamond
85%
Mathematics
MATH-500
87%
Programming
HumanEval
88%
SWE-bench Verified
80.2%
Reasoning
IFEval
85%
Recommended Use Cases
CodingAnalysisResearchData Extraction
Strengths
- Top-tier software engineering (SWE-bench 80.2%, rivaling Claude Opus)
- Extremely cost-efficient — frontier coding at $0.30/$1.20 per M tokens
- Fast inference at 100 TPS with 205K context window
- Open-weight MoE model enabling self-hosting
Limitations
- Weaker on general reasoning and math than top proprietary models
- Smaller community and ecosystem compared to Claude/GPT/Gemini
- Creative writing and conversational fluency lag behind chat-optimized models
Resources
This model may use your data for training
