Back

OpenAI

GPT-6 Luna

ReasoningClosed weights
Intelligence
34.6
Rank 92 of 744 measured models
Output speed
111 t/s
Rank 154 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
$0.0422
Rank 32 of 193 measured models

GPT-6 Luna (Xhigh) scores 34.6 on the Artificial Analysis Intelligence Index, ranking 92nd of 744 measured models. The closest models on this index are Agnes 2.5 Pro Beta (35.2), DeepSeek V4 Flash Vision (Max) (34.8), and DeepSeek V4 Flash 0731 (Max) (34.3). At $0.20 per million tokens (3:1 input/output blended price), it is cheaper than DeepSeek V4 Flash Vision (Max) ($0.66) and DeepSeek V4 Flash 0731 (Max) ($0.66), more expensive than Agnes 2.5 Pro Beta ($0.15). With an output speed of 111 t/s, it is slower than Agnes 2.5 Pro Beta (120 t/s), DeepSeek V4 Flash Vision (Max) (173 t/s), and DeepSeek V4 Flash 0731 (Max) (173 t/s).

Artificial Analysis Indices
Extra high
Intelligence34.6
Rank 92 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 & Accounting37.2
Rank 93 of 218 measured models
Strategy & Ops42.6
Rank 77 of 218 measured models
Legal37.2
Rank 90 of 218 measured models
Engineering33.7
Rank 98 of 217 measured models
Economics41.6
Rank 101 of 225 measured models
Standard Benchmarks
HLE34.3%
Rank 126 of 664 measured models
SciCode51.7%
Rank 88 of 231 measured models
AA-LCR80.0%
Rank 77 of 575 measured models
Terminal-Bench v48.1%
Rank 91 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