readable AI benchmarks (simplified)

readable AI benchmarks

MiniMax

MiniMax-M2.5

reasoning · open weights · Feb 12, 2026

MiniMax-M2.5 has a Quality Score of 32.8, ranking 233rd among 551 scored models. Per 1M tokens, pricing is $0.300 input, $1.20 output, $0.030 cached input, and $0.375 cache write. It accepts text input, outputs text, and has a 204,800 token context window. Its Reliability Score is 30.6, ranking 300th among 553 scored models, below average (total average: 35.6). Its Value Score is 27.6, around average compared with the total Value average of 30.7.

Quality Score32.8coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score27.6
Factual reliability30.6
Cache Discount90%

Model specification

Reasoning
reasoning
Input modalities
text
Output modalities
text
Context window
204,800 tokens
Weights
open weights

Token prices USD per 1M tokens

Input
$0.300
Output
$1.20
Cached Input
$0.030
Cache write
$0.375
Cost per task
not available

Capability

Intelligence
22.8
Coding
not available
Agentic
not available
Omniscience
-38.9
Correct
26.2%
Blended price
$0.201

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 91.2%

Correct 26.2%Incorrect 65.0%Partial / not attempted 8.8%

Artificial Analysis benchmarks

Humanity’s Last Exam
21%
GPQA Diamond
85%
CritPt
1%
AA-Omniscience
31%
AA-LCR
73%

Similar models

Support me! Patreon