MiniMax-M2.5

Creator

MiniMax

Released

2026-02-12

Intelligence

33.7

Artificial Analysis Index

Coding

Artificial Analysis Index

In $/1M

$0.30

input tokens

Out $/1M

$1.20

output tokens

Blended $/1M

$0.53

3:1 blended

Speed

0

tokens / sec

Profile

License, openness, and modality — the governance layer.

License

MIT

commercial OK

Weights

Open

downloadable

Modalities

Image · Text

Parameters

230B

total

MiniMax M2.5 is the world's first production-level model designed natively for Agent scenarios. Building on the M2.1 foundation, M2.5 delivers significant improvements in programming, tool calling, search, and office productivity. With only 10B activation parameters from its 230B MoE architecture, it achieves competitive performance against top international models like Claude Opus 4.6 while maintaining high throughput and efficient inference. M2.5 supports full-stack development for PC, App, and cross-platform applications, and excels in agentic workflows including automated customer support, data-analysis pipelines, and complex task execution.

Capability profile

Category strength across 20 domains, via LLM Stats.

search
80frontend development
80agents
70code
60
finance
60general
60reasoning
60multimodal
60
Via LLM Stats

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
85
Humanity’s Last Exam
19
MMLU-Pro
SciCode
43
LiveCodeBench
MATH-500
AIME 2025
τ²-Bench (agentic)
95
Terminal-Bench Hard
35
IFBench
72
Via Artificial Analysis

Providers

Inference hosts serving this model — their own pricing and measured performance, cheapest input first.

ProviderIn $/1MOut $/1MThroughputLatencyStatus
MiniMax$0.30$1.201003.00active
All providers

1 host · via LLM Stats

© 2026 NYSGPT2525 LLC