@ArizePhoenix: Anthropic's new server-side fallback (beta) is a good case study. Claude Fable 5 and Opus 5 ship with safety classifier…

X AI KOLs Following News

Summary

Anthropic introduces a beta server-side fallback feature for its Claude models, Fable 5 and Opus 5, which uses safety classifiers and automatic API retries to enhance reliability.

Anthropic's new server-side fallback (beta) is a good case study. Claude Fable 5 and Opus 5 ship with safety classifiers that can decline a request. With `fallbacks` enabled, the API automatically retries on a different model and returns one response.
Original Article

Similar Articles

Anthropic launches Opus 5

TechCrunch AI

Anthropic released Opus 5, a new version of its heavyweight model that is cheaper and less restrictive than Fable 5, and outperforms it on some benchmarks. The model also introduces lighter safeguards and a new Automatic Fallbacks feature for API users.

Anthropic’s Claude Fable 5 is a version of Mythos the public can access today

TechCrunch AI

Anthropic launched Claude Fable 5, a publicly accessible version of its powerful Mythos model, with safety guardrails that block responses in high-risk areas and fall back to a weaker model. The release follows Anthropic's warning about AI becoming too dangerous and its push for coordinated safety measures.