正文 Markdown
Claude Fable 5 scores very well on FrontierMath: Tiers 1–4 (v2), reaching 87% on Tiers 1–3 and 88% on Tier 4. This continues a streak of Anthropic models improving rapidly at math.
原文整理页
中文导读
Claude Fable 5 在 FrontierMath 基准测试中表现惊人,Tier 4 准确率达到 88%,标志着 Anthropic 模型数学能力的又一次飞跃。
正文 Markdown
Claude Fable 5 scores very well on FrontierMath: Tiers 1–4 (v2), reaching 87% on Tiers 1–3 and 88% on Tier 4. This continues a streak of Anthropic models improving rapidly at math.