Models / GPT-6 Luna

OpenAI

GPT-6 Luna

33 published results from 4 sources. Each card shows where the number comes from and what it does not measure. Results are never combined into one score.

Provider
OpenAI
Sources
4
Our benchmarks
3
Price
Not yet published

Measured by Spring Prompt

Measured by Spring Prompt · BulletBench
557
Ladder Elo · rank 2 of 16
Unit
ladder Elo, higher is better
Range
370 to 727
Sample
8 games
Configuration
GPT-6 Luna, provider default reasoning
Measured
29 Sep 2026
Not shown
Not a business skill: chess against an engine on a real clock, used here as a speed-under-pressure signal.
Measured by Spring Prompt · BulletBench
7.0 s
Median move time · rank 10 of 16
Unit
milliseconds, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
29 Sep 2026
Not shown
Not general response speed: replies are one short move.
Measured by Spring Prompt · CatalogBench
97.6%
Content quality · rank 6 of 18
Unit
% of checks, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
0.0%
Failed outputs · rank 1 of 18
Unit
% of products, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not a provider outage: provider errors are retried before a product counts as failed.
Measured by Spring Prompt · CatalogBench
61.9%
Publish-ready listings · rank 4 of 18
Unit
% of products, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
42.9%
Reliably publish-ready · rank 4 of 18
Unit
% of products, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
0.56
Unsupported claims · rank 4 of 18
Unit
claims per product, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
100.0%
Channel compliance · rank 1 of 18
Unit
% of products, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
100.0%
Conflicts caught · rank 1 of 18
Unit
% of conflicts, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
97.2%
Content quality · rank 5 of 18
Unit
% of checks, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
$0.0005
Cost per product · rank 1 of 18
Unit
US dollars, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your cost: prices are those charged through OpenRouter on the run date.
Measured by Spring Prompt · CatalogBench
96.7%
Decision accuracy · rank 4 of 18
Unit
% of decisions, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
0.0%
Failed outputs · rank 1 of 18
Unit
% of products, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not a provider outage: provider errors are retried before a product counts as failed.
Measured by Spring Prompt · CatalogBench
93.7%
Field accuracy · rank 7 of 18
Unit
% of missing fields, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
1.7%
Invented values · rank 11 of 18
Unit
% of filled values, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
64.3%
Publish-ready listings · rank 4 of 18
Unit
% of products, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
51.8%
Reliably publish-ready · rank 4 of 18
Unit
% of products, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · CatalogBench
0.41
Unsupported claims · rank 4 of 18
Unit
claims per product, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
30 Sep 2026
Not shown
Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
Measured by Spring Prompt · ROASBench
$0.0142
Cost of a run · rank 1 of 17
Unit
US dollars, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
29 Sep 2026
Not shown
Not your cost: prices are those charged through OpenRouter on the run date.
Measured by Spring Prompt · ROASBench
0
Months over budget · rank 1 of 17
Unit
months of 12, lower is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
29 Sep 2026
Not shown
Not real ad performance: the market is a deterministic simulation of one invented skincare brand.
Measured by Spring Prompt · ROASBench
53.4
Overall score · rank 3 of 17
Unit
score out of 100, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
29 Sep 2026
Not shown
Not real ad performance: the market is a deterministic simulation of one invented skincare brand.
Measured by Spring Prompt · ROASBench
£0.66
Return on ad spend · rank 4 of 17
Unit
profit per £1 spent, higher is better
Configuration
GPT-6 Luna, provider default reasoning
Measured
29 Sep 2026
Not shown
Not real ad performance: the market is a deterministic simulation of one invented skincare brand.

Reported by others

Reported by Arena (formerly LMArena)
0.05
Confirmed task success · rank 5 of 46
Unit
IPS effect estimate, higher is better
Range
0.01 to 0.08
Sample
3913 observations
Configuration
GPT-6 Luna
Measured
28 Sep 2026
Not shown
Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
Reported by Arena (formerly LMArena)
-0.03
Praise over complaint · rank 15 of 46
Unit
IPS effect estimate, higher is better
Range
-0.09 to 0.03
Sample
1083 observations
Configuration
GPT-6 Luna
Measured
28 Sep 2026
Not shown
Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
Reported by Arena (formerly LMArena)
0.02
Steerability · rank 6 of 46
Unit
IPS effect estimate, higher is better
Range
-0.02 to 0.06
Sample
3522 observations
Configuration
GPT-6 Luna
Measured
28 Sep 2026
Not shown
Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
Reported by Arena (formerly LMArena)
0.00
Tool grounding · rank 1 of 46
Unit
IPS effect estimate, higher is better
Range
0.00 to 0.00
Sample
802714 observations
Configuration
GPT-6 Luna
Measured
28 Sep 2026
Not shown
Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
Reported by Arena (formerly LMArena)
1,422
Overall · rank 103 of 177
Unit
Arena rating, higher is better
Range
1,414 to 1,430
Sample
4360 votes
Configuration
GPT-6 Luna (max)
Measured
25 Sep 2026
Not shown
Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.
Reported by Arena (formerly LMArena)
1,452
Business, management and finance · rank 22 of 402
Unit
Arena rating, higher is better
Range
1,432 to 1,473
Sample
875 votes
Configuration
GPT-6 Luna (max)
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,403
Creative writing · rank 68 of 407
Unit
Arena rating, higher is better
Range
1,382 to 1,423
Sample
939 votes
Configuration
GPT-6 Luna (max)
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,505
Expert prompts · rank 4 of 359
Unit
Arena rating, higher is better
Range
1,480 to 1,529
Sample
578 votes
Configuration
GPT-6 Luna (max)
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,443
Instruction following · rank 40 of 409
Unit
Arena rating, higher is better
Range
1,429 to 1,457
Sample
1706 votes
Configuration
GPT-6 Luna (max)
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,442
Overall · rank 67 of 409
Unit
Arena rating, higher is better
Range
1,434 to 1,451
Sample
4792 votes
Configuration
GPT-6 Luna (max)
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,415
Writing, literature and language · rank 67 of 408
Unit
Arena rating, higher is better
Range
1,398 to 1,433
Sample
1245 votes
Configuration
GPT-6 Luna (max)
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.

Compare GPT-6 Luna with