readable AI benchmarks (simplified)

readable AI benchmarks

OpenAI

GPT-6 Luna (Non-reasoning)

non-reasoning · closed weights · Sep 22, 2026

GPT-6 Luna (Non-reasoning) has a Quality Score of 28.3, ranking 107th 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 has a 1M token context window. Its Reliability Score is 39.0, ranking 128th among 282 scored models, above average (total average: 36.9). Its Value Score is 24.9, low compared with the total Value average of 30.3.

Quality Score28.3coverage 78.8%: missing Epoch General ECI, Coding, CursorBench 3.2, Agentic and DeepSWE v1.1
Value Score24.9
Reliability39.0
Cache Discount90%

Model specification

Reasoning
non-reasoning
Input modalities
none
Output modalities
none
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
$0.011

Capability

Intelligence
18.3
Coding
not available
Agentic
not available
Omniscience
-21.9
Correct
31.9%
Blended price
$0.077

Answer outcomes

Attempt rate: 85.6%

Correct 31.9%Incorrect 53.8%Abstained 14.4%

Artificial Analysis benchmarks

GDPval-AA v2
27%
SciCode
43%
Humanity’s Last Exam
9%
CritPt
1%
AA-Omniscience
39%
AA-LCR
40%

Similar models

Support me! Ko-Fi