readable AI benchmarks (simplified)

readable AI benchmarks

OpenAI

GPT-5 Codex (high)

reasoning · closed weights · Sep 23, 2025

GPT-5 Codex (high) has a Quality Score of 42.3, ranking 154th among 551 scored models. Per 1M tokens, pricing is $1.25 input, $10.00 output, and $0.125 cached input. It accepts text and image input, outputs text, and has a 400k token context window. Its Reliability Score is 46.2, ranking 151st among 553 scored models, above average (total average: 35.6). Its Value Score is 28.5, around average compared with the total Value average of 30.7.

Quality Score42.3coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score28.5
Factual reliability46.2
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
$1.25
Output
$10.00
Cached Input
$0.125
Cache write
not available
Cost per task
not available

Capability

Intelligence
24.9
Coding
not available
Agentic
not available
Omniscience
-7.5
Correct
38.7%
Blended price
$1.34

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 84.9%

Correct 38.7%Incorrect 46.2%Partial / not attempted 15.1%

Artificial Analysis benchmarks

Humanity’s Last Exam
28%
GPQA Diamond
84%
CritPt
5%
AA-Omniscience
46%
AA-LCR
72%

Similar models

Support me! Patreon