readable AI benchmarks (simplified)

readable AI benchmarks

MiniMax

MiniMax-M2.7

reasoning · open weights · Mar 18, 2026

MiniMax-M2.7 has a Quality Score of 42.7, ranking 148th among 551 scored models. Per 1M tokens, pricing is $0.300 input, $1.20 output, $0.060 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 50.4, ranking 108th among 553 scored models, above average (total average: 35.6). Its Value Score is 35.6, high compared with the total Value average of 30.7.

Quality Score42.7coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score35.6
Factual reliability50.4
Cache Discount80%

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
$0.060
Cache write
$0.375
Cost per task
$0.102

Capability

Intelligence
22.8
Coding
not available
Agentic
not available
Omniscience
0.8
Correct
26.8%
Blended price
$0.222

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 52.8%

Correct 26.8%Incorrect 26.0%Partial / not attempted 47.2%

Artificial Analysis benchmarks

GDPval-AA v2
25%
τ³-Banking
10%
SciCode
50%
Humanity’s Last Exam
30%
GPQA Diamond
87%
CritPt
1%
AA-Omniscience
50%
AA-LCR
78%

Similar models

Support me! Patreon