Source Intelligence

DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.

All of v2.1.225 Home All releases olderv2.1.224 v2.1.226newer

Auto-mode classifier now falls back when the primary model refuses

Not switched on
Useful2 Signal2
Elsewhere

When the auto-mode model refuses a request, a backup model now gets a try instead of failing.

Fallback-on-refusal logic is present in the auto-mode classifier, but its rollout state couldn't be resolved from the build.

What

A safeguard refusal from the primary classifier model is treated like the model being unavailable, so the fallback model gets a chance instead of the request failing outright. The rollout state of this path could not be resolved in this build.

Details
  • refusedBySafeguard joins unavailability as a trigger for retrying on the fallback model within the remaining time budget
  • Usage and durations from both attempts are merged
  • If the fallback also fails to produce a usable answer, the original result is kept
  • The log line now distinguishes a refusal from an unavailability
Evidence

refused (stop_reason=refusal)

Strings lifted out of the shipped bundle, so the claim above can be checked against them.

See this entry in the whole of v2.1.225 →