readable AI benchmarks (simplified)

readable AI benchmarks

OpenAI

GPT-6 Luna (max)

reasoning · closed weights · Sep 22, 2026

GPT-6 Luna (max) has a Quality Score of 49.5, ranking 55th among 282 scored models. Per 1M tokens, pricing is $0.100 input, $0.500 output, $0.010 cached input, and $0.125 cache write. It accepts text, image, and pdf input, outputs text, and has a 1M token context window. Its Reliability Score is 50.3, ranking 64th among 282 scored models, above average (total average: 36.9). Its Value Score is 45.0, high compared with the total Value average of 30.3.

Quality Score49.5coverage 78.8%: missing Epoch General ECI, Coding, CursorBench 3.2, Agentic and DeepSWE v1.1
Value Score45.0
Reliability50.3
Cache Discount90%

Model specification

Reasoning
reasoning
Input modalities
text, image, pdf
Output modalities
text
Context window
1M tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$0.100
Output
$0.500
Cached Input
$0.010
Cache write
$0.125
Cost per task
$0.068

Capability

Intelligence
37.3
Coding
not available
Agentic
not available
Omniscience
0.7
Correct
43.8%
Blended price
$0.077

Answer outcomes

Attempt rate: 86.9%

Correct 43.8%Incorrect 43.1%Abstained 13.1%

Artificial Analysis benchmarks

GDPval-AA v2
43%
SciCode
55%
Humanity’s Last Exam
39%
CritPt
19%
AA-Omniscience
50%
AA-LCR
83%

Similar models

Support me! Ko-Fi