MiniMax: MiniMax M2.5
Analysis Summary
MiniMax M2.5 is a strong mid-tier model with an intelligence index of 41.9, a coding index of 37.4, and an agentic index of 65.1. Its instruction following score of 0.716 is among the highest in its tier, and long-context reliability at 0.66 is solid. Tool use and function calling are supported, and the tau2 score of 0.953 indicates very strong multi-step task completion.
For businesses, M2.5 is well-suited to agentic pipelines, structured content generation, and high-volume automation where instruction precision matters. The 204K context window is adequate for most document tasks. It does not support vision, which limits its use in multimodal workflows. Coding capability is reasonable but not class-leading.
At $0.15 input and $0.90 output per million tokens, it is aggressively priced for its capability level. Teams running cost-sensitive agentic or content workflows will find strong value, though businesses requiring vision support or frontier reasoning should look to higher-tier options.
Assessed June 9, 2026
Editorial notes
MiniMax M2.5 combines a strong agentic index of 65.1 with excellent instruction following and competitive pricing, making it a capable choice for agentic and content workflows at scale.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How MiniMax: MiniMax M2.5 compares
MiniMax: MiniMax M2.5 ranks #49 of 378 AI models we track for overall intelligence, #49 of 315 for coding, #35 of 289 for agentic tasks. Its 205K-token context window is larger than 68% of the models we list. At $0.15 per million input tokens it is cheaper than 57% of comparable models.
About MiniMax: MiniMax M2.5
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..
Capabilities
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does MiniMax: MiniMax M2.5 stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
minimax/minimax-m2.5
|
| Provider | minimax |
| Release Date | February 12, 2026 |
| Context Length | 204,800 tokens |
| Max Completion | 196,608 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.15 | $0.000150 |
| Output | $0.90 | $0.000900 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about MiniMax: MiniMax M2.5
How much does MiniMax: MiniMax M2.5 cost?
MiniMax: MiniMax M2.5 costs $0.15 per million input tokens and $0.90 per million output tokens.
What is the context window of MiniMax: MiniMax M2.5?
MiniMax: MiniMax M2.5 has a context window of 204,800 tokens (205K).
Is MiniMax: MiniMax M2.5 good for coding?
On our coding benchmark index, MiniMax: MiniMax M2.5 ranks #49 of 315 models, placing it in the top quartile of the field for code generation and debugging.
What can MiniMax: MiniMax M2.5 do?
MiniMax: MiniMax M2.5 supports tool use and function calling.
Who created MiniMax: MiniMax M2.5?
MiniMax: MiniMax M2.5 is developed by MiniMax and was released on February 12, 2026.
