YOU ARE VIEWING ONE ITEM FROM THE AICRIER FEED

Claude Sonnet 5.5 Rejects Disabled Thinking

AICrier tracks AI developer news across Product Hunt, GitHub, Hacker News, YouTube, X, arXiv, and more. This page keeps the article you opened front and center while giving you a path into the live feed.

// WHAT AICRIER DOES

7+

TRACKED FEEDS

24/7

SCRAPED FEED

Short summaries, external links, screenshots, relevance scoring, tags, and featured picks for AI builders.

Claude Sonnet 5.5 Rejects Disabled Thinking
OPEN LINK ↗
// 1h agoPRODUCT UPDATE

Claude Sonnet 5.5 Rejects Disabled Thinking

Anthropic’s Claude Sonnet 5.5 returns a 400 error when requests use `thinking: {"type": "disabled"}`. Developers migrating from Sonnet 5 must use `thinking: {"type": "between_tools"}` to suppress upfront thinking.

// ANALYSIS

This is a small parameter change with outsized migration risk for production agent stacks.

  • –`between_tools` is the lowest thinking mode and works at low, medium, and high effort.
  • –It fails at xhigh or max effort, where adaptive thinking must be used instead.
  • –Tool-driven applications may now receive progress updates inside `thinking` blocks.
  • –Existing integrations should update configuration before switching model IDs to avoid runtime failures.
// TAGS
claude-sonnet-5-5llmapireasoningtool-usedevtool

DISCOVERED

1h ago

2026-09-29

PUBLISHED

1h ago

2026-09-29

RELEVANCE

9/ 10

AUTHOR

cjav_dev