You'll noticeTier: how much it should matter to you
3Useful: my rating, 1 to 5
1Signal: worth watching, 1 to 5
Model FallbackArea: what it touches
ImprovementsKind: in v2.1.286,
ImprovementsSection of the release
What
Claude Code can switch to a fallback model, a backup from your fallback list, when the model you picked cannot be used. Each model has a context window, which is how much of the conversation it can hold at once. When the fallback model has a smaller context window than the one it replaces, Claude Code now says so.
The notice gains a line showing the change, in the form "context window X → Y tokens".
A follow-up message explains that Claude Code fell back to another model and suggests choosing a fallback model with a larger context window, or using /clear to start a new conversation with empty context.
It also suggests putting a model that supports a 1M-token context window in the fallback list.
Why
Without this, a long conversation could suddenly seem to lose earlier details after a fallback, with no hint why. The notice explains the loss and points to ways to avoid it.
How sure we are
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubtIt is not clear whether this notice is shown to everyone or only when some setting allows it.
Anthropic's release notes agreeFixed /compact, /clear, and /rewind typed while viewing a background agent's or teammate's transcript silently acting on the main…
