Skip to content
Trending storyDeveloping

Arena ranks Claude Opus 5.5 (Max) first on Image-to-WebDev Arena, Sonnet 5.5 second

1 article1 sourcesince Oct 10Last article 1h ago ·

Overview

AISummary of 1 article

Arena says Claude Opus 5.5 (Max) ranks first on its Image-to-WebDev Arena with 1749 points, and Claude Sonnet 5.5 (xHigh) ranks second with 1740 points.

Both models sit on the Pareto frontier, Arena says.

Arena lists Opus 5.5 at $16 per million blended input/output tokens and Sonnet 5.5 at $8 per million, 50% less than Opus for a 9-point score gap. It puts GPT-6 Astra at $40 per million, 60% above Opus's price. Arena says it will evaluate Claude Haiku 5.5 next.

Written by AI from the articles below · updated Oct 10, 1:26 PM ET

Check the sources:

Article timeline

The articles in this story. Times are ET.

Oct 10
  1. Arena.aiOfficial
    Claude Opus 5.5 tops Image-to-WebDev Arena, Sonnet 5.5 close behind

    AIArena says Claude Opus 5.5 (Max) ranks first on its Image-to-WebDev Arena with 1749 points, and Claude Sonnet 5.5 (xHigh) ranks second with 1740 points. Both models sit on the Pareto frontier, and Opus 5.5 costs $16 per million blended input/output tokens, 60% less than GPT-6 Astra's $40 per million. Sonnet 5.5 costs $8 per million tokens, 50% less than Opus, and Arena says Claude Haiku 5.5 will be evaluated next.

    Image from @arena's post

Heat trend

Not enough continuous observations to show a trend yet.