Claude Fable 5.1 thinking blocks only replay to the same or newer models
UpdateVerifiedAdded Sep 22, 2026
Thinking blocks from Claude Fable 5.1 and Mythos 5.1 can't be read by earlier models and are dropped if replayed to them. Fable 5.1 also checks that nothing before a replayed block has changed, which matters for agents that edit conversation history.
Topics: LLMs, Developer tools
Summaries of vendors' own notes. Product names and logos belong to their owners; logos via logo.dev.
More Claude Platform releases
Every Claude Platform release| Date | Release | Type |
|---|---|---|
| Sep 1 | Claude Fable 5.1 launches alongside Claude Mythos 5.1 with 1M token context GA | GA |
| Sep 1 | Prompt cache reads on Claude Fable 5.1 and Mythos 5.1 drop to $0.25 per MTok Pricing | Pricing |
| Sep 1 | Claude Fable 5.1 and Mythos 5.1 reject tool_choice any and tool Update | Update |
| Sep 1 | Per-message effort changes enter beta on the Claude API Beta | Beta |
| Sep 1 | Claude API adds turn-scoped system messages in beta Beta | Beta |
| Sep 1 | Claude API thinking.display adds an updates mode in beta Beta | Beta |
| Sep 1 | Claude Fable 5.1 output carries text watermarks and C2PA Content Credentials Update | Update |
| Sep 1 | Claude Fable 5.1 and Mythos 5.1 require 30-day data retention Update | Update |
