Trending storyDeveloping

ARC Prize reports verified ARC-AGI scores for Claude Sonnet 5.5

1 article1 sourcesince Oct 21Last article 2h ago · Oct 21, 3:30 PM ET

Overview

AISummary of 1 article

ARC Prize reports that Claude Sonnet 5.5 from Anthropic scores 72.4% on ARC-AGI-3 at a cost of $8.2K using the provider adapter harness.

On the standard harness, the same model scores 24.3% at $7.4K, according to the same report. ARC Prize also reports earlier-version scores of 83.9% on ARC-AGI-2 at $0.55 per task and 96.5% on ARC-AGI-1 at $0.14 per task.

The results are posted on X under ARC Prize's account, with the evidence line describing them as verified results from Anthropic's account.

Written by AI from the articles below · updated Oct 21, 3:35 PM ET

Check the sources:

Latest developmentOct 21, 3:30 PM ET

Claude Sonnet 5.5 scores 72.4% on ARC-AGI-3 with provider adapter harness

Article timeline

The articles in this story. Times are ET.

Oct 21

3:30 PM
ARC Prize@arcprizeOfficial
Claude Sonnet 5.5 scores 72.4% on ARC-AGI-3 with provider adapter harness
AIARC Prize reports that Claude Sonnet 5.5 from Anthropic scores 72.4% on ARC-AGI-3 at a cost of $8.2K using the provider adapter harness, versus 24.3% at $7.4K with the standard harness. On earlier versions, it scores 83.9% on ARC-AGI-2 at $0.55 per task and 96.5% on ARC-AGI-1 at $0.14 per task.

Heat Trend

Not enough continuous observations to show a trend yet.