Back

OpenAI

GPT-6 Luna

ReasoningClosed weights
Intelligence
29.9
Rank 129 of 744 measured models
Output speed
143 t/s
Rank 100 of 403 measured models
Price/1M
$0.20
Rank 161 of 659 measured models
Context window
1M
Rank 84 of 1199 measured models
Cost per task
$0.0175
Rank 20 of 193 measured models

GPT-6 Luna (Medium) scores 29.9 on the Artificial Analysis Intelligence Index, ranking 129th of 744 measured models. The closest models on this index are Claude Sonnet 4.6 (batch) (30.1), Gemini 3.1 Pro Preview (29.7), and Qwen3.7 Max (29.5). At $0.20 per million tokens (3:1 input/output blended price), it is cheaper than Gemini 3.1 Pro Preview ($4.50) and Qwen3.7 Max ($3.75). With an output speed of 143 t/s, it is faster than Gemini 3.1 Pro Preview (131 t/s), slower than Qwen3.7 Max (191 t/s).

Artificial Analysis Indices
Medium
Intelligence29.9
Rank 129 of 744 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Cost per task
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting32.9
Rank 106 of 218 measured models
Strategy & Ops37.5
Rank 96 of 218 measured models
Legal33.0
Rank 108 of 218 measured models
Engineering29.4
Rank 115 of 217 measured models
Economics37.1
Rank 121 of 225 measured models
Decision Index
GPT-6 Luna (Medium) · Artificial Analysis
Decision Index58.2Rank 12 of 26 measured models
AA-Omniscience-Decision · accuracy71.2
AA-Omniscience-Decision · non-hallucination0.0
HLE-MultiChoice29.0
When2Call78.2
WildGuardTest89.8
Cost per 1M tasks$181.79
Average time per task1.73 s
Decision Index leaderboard
Standard Benchmarks
HLE28.3%
Rank 166 of 664 measured models
SciCode50.9%
Rank 99 of 231 measured models
AA-LCR78.3%
Rank 111 of 575 measured models
Terminal-Bench v42.5%
Rank 109 of 218 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy