readable AI benchmarks (simplified)

readable AI benchmarks

OpenAI

GPT-5 mini (high)

reasoning · closed weights · Aug 7, 2025

GPT-5 mini (high) has a Quality Score of 33.6, ranking 221st among 551 scored models. Per 1M tokens, pricing is $0.250 input, $2.00 output, and $0.025 cached input. It accepts text and image input, outputs text, and has a 400k token context window. Its Reliability Score is 41.3, ranking 205th among 553 scored models, above average (total average: 35.6). Its Value Score is 27.4, around average compared with the total Value average of 30.7.

Quality Score33.6coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score27.4
Factual reliability41.3
Cache Discount90%

Model specification

Reasoning
reasoning
Input modalities
text, image
Output modalities
text
Context window
400k tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$0.250
Output
$2.00
Cached Input
$0.025
Cache write
not available
Cost per task
$0.053

Capability

Intelligence
16.8
Coding
not available
Agentic
not available
Omniscience
-17.3
Correct
25.0%
Blended price
$0.267

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 67.3%

Correct 25.0%Incorrect 42.3%Partial / not attempted 32.7%

Artificial Analysis benchmarks

GDPval-AA v2
13%
τ³-Banking
15%
SciCode
39%
Humanity’s Last Exam
21%
GPQA Diamond
83%
CritPt
0%
AA-Omniscience
41%
AA-LCR
72%

Similar models

Support me! Patreon