Skip to content
View original post on X: ARC PrizeOfficial
Published
46/100

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.
ARC PrizeVerified on XOfficial
@arcprize
View on X

Claude Sonnet 5.5 from @AnthropicAI on ARC-AGI (Verified):

  • ARC-AGI-3: 72.4%, $8.2K (provider adapter harness), 24.3%, $7.4K (standard harness)
  • ARC-AGI-2: 83.9%, $0.55/task
  • ARC-AGI-1: 96.5%, $0.14/task

Sonnet 5.5 is notable for its 72.4% score on ARC-AGI-3.

Source details

ARC Prize · x.comPublished

Original title: Claude Sonnet 5.5 from @AnthropicAI on ARC-AGI (Verified):

Topics

Anthropic / ClaudeReasoningEvals & benchmarks