modelgrep
M

MiniMax: MiniMax M2.5

minimax/minimax-m2.5

37th smartest of 180ReasoningToolsJSON
Use via OpenRouter ↗
Intelligence
41.9
37th of 180
Design Elo
1266
Website
Speed
199
17th fastest
Latency
521ms
first token
Input price
$0.150
79th cheapest
Context
205K
197K max out

How it compares

Smarter than79%
of all ranked models
Faster than94%
of all ranked models
Cheaper than73%
of all ranked models

Overview

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...

Benchmarks

independent · via OpenRouter
Artificial Analysis77th percentile
Intelligence Index
41.9
Coding Index
37.4
Agentic Index
55.6
GPQA Diamond
85%
Humanity's Last Exam
19%
SciCode
43%
Tau²-Bench (agentic)
95%
Design Arena · Elo5,389 tournaments
Website
1266
codecategories
1258
3D
1251
Game Dev
1245
UI Component
1229
Data Viz
1216
svg
1208

Providers & pricing (17)

ProviderIn $/MOut $/MUptime
Inceptronfp8$0.150$0.90099.9%
AkashMLfp8$0.150$1.1599.6%
DeepInfrafp8$0.150$1.15100%
Chutesfp8$0.150$1.20100%
Phala$0.200$1.38100%
DigitalOcean$0.225$0.900100%
StreamLake$0.270$1.08
AtlasCloudfp8$0.295$1.20
Mara$0.300$1.20
Friendli$0.300$1.20100%
Minimaxfp8$0.300$1.20100%
Novitafp8$0.300$1.20
SiliconFlowfp8$0.300$1.20
Parasailfp8$0.300$1.2099.3%
WandBfp8$0.300$1.20
Venice$0.340$1.19
Minimaxfp8$0.600$2.40

Specifications

Context window205K
Max output197K
Knowledge cutoff
Input modalitiestext
Output modalitiestext
Prompt caching
Cache read price$0.050/M
ModeratedNo

MiniMax M2.5 FAQ

How much does MiniMax M2.5 cost?

MiniMax M2.5 costs $0.150 per million input tokens and $0.900 per million output tokens via OpenRouter, making it 79th cheapest of 298 paid models.

How smart is MiniMax M2.5?

MiniMax M2.5 scores 41.9 on the Artificial Analysis Intelligence Index, ranking 37th of 180 benchmarked models, with a GPQA Diamond score of 85%.

How fast is MiniMax M2.5?

MiniMax M2.5 generates around 199 tokens per second with 521ms time-to-first-token (p50), the 17th fastest tracked model.

What is MiniMax M2.5's context window?

MiniMax M2.5 supports a 205K-token context window and can output up to 197K tokens. It accepts text input.

Compare head-to-head