Sonnet 5.5 keeps $2/$10 per million tokens; thinking always on, `between_tools` disables it
ClaudeDevs · x · 2026-09-29
Claude Sonnet 5.5 is priced the same as Sonnet 5: $2/$10 per million tokens with cache reads at $0.20.
Migration guide highlights: thinking now runs by default (adaptive mode), and you can use betweentools to turn off upfront thinking. Five legacy settings return 400 errors: thinking budgets, sampling parameters, assistant prefill, forced tool choice, and thinking: {"type": "disabled"}. Anthropic provides per-model migration checklists from Sonnet 5/4.6/4.5/4/3.7 and Haiku 4.5.
Related event: Anthropic Launches Claude Sonnet 5.5: 30% Faster, Up to 30% Cheaper(35 posts)→
More from coding & agent
- Every team tests Sonnet 5.5: half Opus's token price, matches it on writing readability — kieranklaassen · 2026-09-29
- Giving a deployment agent persistent memory with Hindsight — Kshitiii · 2026-09-29
- Four one-time Claude Skills (about-me, anti-ai-writing, skill-creator) automate your whole workflow — alex_verem · 2026-09-29
- Dev builds no-login pickleball court aggregator with agents, 500 visitors in 24 hours — AJChadha · 2026-09-29
- TypeScript C++ rewrite on track to be ~10x faster than Go version — inductionheads · 2026-09-29
- Grok launches Team Bots: shared AI teammates that learn as your team works — billyuchenlin · 2026-09-29