readable AI benchmarks

OpenAI

GPT-5.6 Luna (medium)

reasoning · closed weights · Jul 9, 2026

Quality Score46.4
Value Score42.0
Reliability43.4
Cache Discount90%

Model specification

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

Token prices USD per 1M tokens

Input
$0.200
Output
$1.20
Cached Input
$0.020
Cached Output
$0.250
Cost per task
$0.011

Capability

Intelligence
38.9
Coding
50.7
Agentic
31.8
Omniscience
-13.2
Correct
40.7%
Blended price
$0.174

Answer outcomes

Correct 40.7%Incorrect 53.9%Abstained 5.4%

Artificial Analysis benchmarks

GDPval-AA v2
39%
τ³-Banking
18%
Terminal-Bench v2.1
53%
SciCode
46%
Humanity’s Last Exam
26%
GPQA Diamond
86%
CritPt
5%
AA-Omniscience
43%
AA-LCR
72%

Similar models