Benchmark menu

Model detail

GPT-5.6 Luna

A static, shareable model page that remains complete without JavaScript.

GPT-5.6 Luna

Provider: OpenAI · Public slug: gpt-5-6-luna

Rank
20
Rank movement
Baseline
Score
32.3
Last measured
2026-07-13
Coverage
Complete coverage

Reasoning scores

XHigh
55.0
Medium
30.2
None
11.6

Generation health

XHigh78%11%11%n 9Medium89%0%11%n 9None88%12%0%n 8 Complete coverage

Cost summary

Mean standard list-price equivalent
$1.07
Median unique-cell cost
$0.04
Unique-cell range
$0.02–$4.99
Coverage
24 unique cells / 26 observed runs
Observed-run total
$29.30
Evidence basis
Standard list-price estimate
Mean output
186k tok
Output range
1.5k–830k tok
Token samples
26

Cost means equal-weight unique model/game/reasoning cells. Provider-account billed spend is reported separately on Charts.

Run setup

  • Overall Models: combines public reasoning variants
  • Standard GameBench code-generation prompt