// AI Platform Drift
Claude Fable 5.1 tool_choice types 'any' and 'tool' return 400 error
Claude Platform release notes · published 2026-11-11 05:31 UTC
WHAT CHANGEDClaude Fable 5.1 and Claude Mythos 5.1 do not support tool_choice types 'any' and 'tool'; requests using them return a 400 error. tool_choice types 'auto' and 'none' are unchanged. For accounts created on or after 31 August 2026, replaying a thinking block after the system prompt, tools, or an earlier message changed returns a 400 error.
WHY IT MATTERSExisting code that sets tool_choice to 'any' or 'tool' will break when migrated to Claude Fable 5.1 or Claude Mythos 5.1. Multi-turn conversations that replay thinking blocks will fail for new accounts if the conversation context changed between turns.
WHAT TO DOGrep your codebase for tool_choice set to 'any' or 'tool' and replace with 'auto', 'none', strict tool use, or structured outputs before migrating to Claude Fable 5.1. If you replay thinking blocks across turns, ensure the system prompt, tools, and message history remain unchanged, or adopt the thinking-binding-controls-2026-08-01 beta header to handle prefix mismatches.
Model-reported confidence 95% (not measured accuracy)
Primary source ↗
written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
https://platform.claude.com/docs/en/release-notes/overview#september-1-2026 Tue, 01 Sep 2026 00:00:00 GMT -We’ve launched Claude Fable 5.1 (claude-fable-5-1), the successor to Claude Fable 5 for long-running agentic coding, knowledge work, and research, alongside Claude Mythos 5.1 (claude-mythos-5-1) for Project Glasswing participants. Both models support a 1M token context window by default, 128k max output tokens, and always-on adaptive thinking, at $10 / $50 USD per MTok, the same as Claude Fable 5, with cache reads cut to $0.25 USD per MTok. Claude Fable 5.1 is available on the Claude API, Claude in Amazon Bedrock, Claude Platform on AWS, Claude on Google Cloud, and Claude in Microsoft Foundry. See What’s new in Claude Fable 5.1 for capabilities, API changes, and migration guidance. +We’ve launched Claude Fable 5.1 (claude-fable-5-1), the successor to Claude Fable 5 for long-running agentic coding, knowledge work, and research, alongside Claude Mythos 5.1 (claude-mythos-5-1) for Project Glasswing participants. Both models support a 1M token context window by default, 128k max output tokens, and always-on adaptive thinking, at $10 / $50 USD per MTok, the same as Claude Fable 5, with cache reads cut to $0.25 USD per MTok. Claude Fable 5.1 is available on the Claude API, Claude in Amazon Bedrock, Claude Platform on AWS, Claude Models on Google Cloud (Google managed), and Claude in Microsoft Foundry. See What’s new in Claude Fable 5.1 for capabilities, API changes, and migration guidance. Prompt cache reads on Claude Fable 5.1 and Claude Mythos 5.1 cost $0.25 USD per million tokens: 0.025x the base input price, compared with 0.1x on other models. Cache writes are unchanged. See Prompt caching pricing. On Claude Fable 5.1 and Claude Mythos 5.1, tool_choice types any and tool aren’t supported and return a 400 error. auto and none are unchanged. To guarantee schema-conformant tool inputs, use strict tool use or structured outputs. Thinking blocks produced by Claude Fable 5.1 and Claude Mythos 5.1 are preserved only for the model that produced them or a newer one: earlier models can’t read them, and the API drops one replayed to an earlier model. Claude Fable 5.1 accepts thinking blocks from Claude Opus 5, Claude Fable 5, Claude Mythos 5, and earlier Claude models. On Claude Fable 5.1, the API also checks that nothing before a block has changed: for new accounts created on or after August 31, 2026, replaying one after the system prompt, tools, or an earlier message changed returns a 400 error. With the thinking-binding-controls-2026-08-01 beta header, dropped blocks are reported in an input_transformations response field, and thinking.block_binding.prefix_mismatch_behavior chooses between rejecting and dropping blocks whose history changed. See Preserved thinking.
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
| triage verdict | MATERIAL at 0.92, needed 0.45 |
|---|---|
| triage reasoning | New model release with pricing, breaking API changes, and account-level behavior changes affecting production integrations. |
| signals | new model release (Claude Fable 5.1, Claude Mythos 5.1) · pricing stated ($10/$50 USD per MTok input/output, $0.25 cache reads) · breaking API change (tool_choice types 'any' and 'tool' return 400 error) · new account behavior change (thinking block replay validation effective Aug 31, 2026) · new beta header and API response field (thinking-binding-controls-2026-08-01) · platform availability change (Google Cloud now 'Google managed') |
| suppression | — |
| triage model | anthropic/claude-haiku-4.5 ($0.00000) |
| writer model | anthropic/claude-sonnet-4.5 ($0.00000) |