做 AI 代理或知识工作自动化的团队值得关注——Claude Fable 5 在真实世界任务基准上登顶，且安全回退机制降低了误判风险，可以直接评估是否适合你的场景。
Anthropic 发布了 Claude Fable 5，这是首个公开可用的 Mythos 级模型，在代理型真实世界知识工作基准测试 GDPval-AA 中排名第一。该模型与 Claude Mythos 5 共享底层模型，但增加了针对网络安全、生物、化学和蒸馏相关有害查询的安全防护。Fable 5 还引入了回退机制，可将标记的查询路由到 Claude Opus 4.8 等第二模型。在 GDPval-AA 上，Fable 5 得分 1932，使 Anthropic 模型占据前四名中的三席。回退发生在不到 5% 的会话中，性能稳定。
Anthropic has released Claude Fable 5, the first p…
Anthropic has released Claude Fable 5, the first publicly available Mythos-class model that ranks #1 in our agentic real-world knowledge work benchmark GDPval-AA
Claude Fable 5 shares the same underlying model as Claude Mythos 5, with added security guardrails for potentially harmful cybersecurity, biology, chemistry, and distillation-related queries. The release also introduces a fallback mechanism, allowing Claude Fable 5 to route flagged queries to a second model such as Claude Opus 4.8.
@AnthropicAI shared access with us ahead of public release to benchmark this model. Claude Fable 5 scores 1932 on GDPval-AA, our benchmark for agentic real-world work tasks, taking the #1 position and putting Anthropic models in 3 of the top 4 spots. The result was measured using adaptive reasoning at max effort, with Claude Opus 4.8 configured as the fallback model. Fable 5 falls back to Opus 4.8 on 2% of GDPval-AA tasks, with Anthropic stating that fallback occurs in fewer than 5% of sessions on average.
Full benchmarks for Claude Fable 5 are in progress - we will share the full Intelligence Index and publish scores on our website shortly