OpenAI

GPT-6.2 Sol

Intelligence
52.7
Rank 27 of 813 measured models
Output speed
63 t/s
Rank 345 of 476 measured models
Context window
1M
Rank 120 of 1420 measured models
Cost per task
$0.406
Rank 105 of 255 measured models

GPT-6.2 Sol (Medium) is a text AI model from OpenAI, released on Nov 17, 2026, with a context window of 1M tokens. GPT-6.2 Sol (Medium) scores 52.7 on the Artificial Analysis Intelligence Index, ranking 27th of 813 measured models. The closest models on this index are Claude Sonnet 5.5 (Default Fallback) (54.5), Gemini 4 Hydrogen (53.9), and GPT-6.1 Terra (51.3). With an output speed of 63 t/s, it is about as fast as GPT-6.1 Terra (66.3 t/s), slower than Claude Sonnet 5.5 (Default Fallback) (151 t/s) and Gemini 4 Hydrogen (152 t/s). Every benchmark side by side: GPT-6.2 Sol (Medium) vs Gemini 4 Hydrogen.

Artificial Analysis Indices

Medium
Intelligence52.7
Rank 27 of 813 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 & Accounting51.6
Rank 43 of 283 measured models
Strategy & Ops56.3
Rank 34 of 283 measured models
Legal48.8
Rank 55 of 283 measured models
Healthcare & Medical48.0
Rank 37 of 210 measured models
Engineering56.2
Rank 34 of 281 measured models
Economics51.1
Rank 52 of 289 measured models

Cyber Index

Artificial Analysis · v1
Cyber Index36.0%Rank 25 of 36 measured models
Successes
36.0%
Safety blocks
34.2%
Failures
29.8%
CWE-Bench-AA62.5% (safety blocks)
DeepsecBench-AA45.5%
CyberGym-E2E-AA0.0% (safety blocks)
Cost per task—

* The model or provider declined some tasks on safety grounds.

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