跳到正文
ARC Prize· @arcprize · X·· 3 小时前AI 评分43
AI 导读

来自 @AnthropicAI 的 Haiku 5.5 在 ARC-AGI(Verified)上的表现: - ARC-AGI-3:2.8%,$437(provider adapter harness);0.5%,$388(standard harness) - ARC-AGI-2:61.3%,$0.085/task - ARC-AGI-1:89.0%,$0.051/task Haiku 5.5 为 Anthropic 的小模型层级带来了有竞争力的性能。

正文

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

  • ARC-AGI-3: 2.8%, $437 (provider adapter harness); 0.5%, $388 (standard harness)
  • ARC-AGI-2: 61.3%, $0.085/task
  • ARC-AGI-1: 89.0%, $0.051/task

Haiku 5.5 delivers competitive performance for Anthropic's small-model tier.

来源:ARC Prize · x.com