Release Date 2026-11-171.1M IN · 128K OUTProprietary ModelAPI availablePerplexity Pro and up
Overall3.5Top 11%
AutomationResearching
Docs & officeResearching
CodingResearching

API cost

Official price
$2IN$10OUT
SourceOpenAI official docs

Output Speed

Slowper 2,000 tokens
54tok/s
First Output 2.7stotal: 39.7s
SourceOpenRouterLast updated

Expert Performance Evaluation

How strong is it across many tests combined?AA Intelligence Index v5Top 10%#5
57.3Reasoning effort: Max

By effortLow 40.6Medium 52.7High 54.9Extra High 55.7Max 57.3

More about AA Intelligence Index v5

24.8%Reasoning effort: Max37.0%12.1%58.9%

More about Tax Agent Bench

Can it carry out scientists' real research tasks in a terminal?Terminal-Bench-Science 0.1Top 8%#3
58.6%Reasoning effort: Max

More about Terminal-Bench-Science 0.1

Can it build a working web app from a plain-language spec?Vibe Code Bench v1.1Top 18%#16
87.6%Reasoning effort: Max

More about Vibe Code Bench v1.1

See each evaluation’s detail page for its source and last updateData corrections & removal requests