OpenAI

GPT-6 Astra

Intelligence
54.8
Rank 11 of 788 measured models
Coding
77.1
Rank 8 of 301 measured models
Output speed
39.8 t/s
Rank 413 of 454 measured models
Price/1M
$20
Rank 717 of 741 measured models
Context window
1M
Rank 107 of 1334 measured models
Cost per task
$3.24
Rank 194 of 235 measured models

GPT-6 Astra (High) scores 54.8 on the Artificial Analysis Intelligence Index, ranking 11th of 788 measured models. The closest models on this index are GPT-6.1 Sol (54.6), Claude Sonnet 5.5 (Default Fallback) (54.5), and Gemini 4 Hydrogen (53.9). At $20 per million tokens (3:1 input/output blended price), it is more expensive than GPT-6.1 Sol ($4) and Claude Sonnet 5.5 (Default Fallback) ($4). With an output speed of 39.8 t/s, it is slower than GPT-6.1 Sol (48.8 t/s), Claude Sonnet 5.5 (Default Fallback) (140 t/s), and Gemini 4 Hydrogen (182 t/s).

Artificial Analysis Indices
High
Intelligence54.8
Rank 11 of 788 measured models
Coding77.1
Rank 8 of 301 measured models
Agentic48.2
Rank 23 of 191 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 & Accounting56.1
Rank 13 of 260 measured models
Strategy & Ops57.2
Rank 19 of 260 measured models
Legal55.1
Rank 14 of 260 measured models
Healthcare & Medical51.6
Rank 12 of 190 measured models
Engineering60.0
Rank 14 of 259 measured models
Economics55.7
Rank 19 of 267 measured models
Standard Benchmarks
GPQA Diamond94.9%
Rank 4 of 621 measured models
HLE53.1%
Rank 18 of 682 measured models
SciCode55.4%
Rank 53 of 249 measured models
AA-LCR80.0%
Rank 86 of 593 measured models
Terminal-Bench 2.189.9%
Rank 3 of 246 measured models
Terminal-Bench 4.054.0%
Rank 18 of 246 measured models
τ-Bench Banking40.0%
Rank 32 of 211 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