• Sources: primary, discussion
  • Summary: Anthropic released Claude Haiku 5.5 on 2026-10-07 as model id claude-haiku-5-5, available on the Claude Platform plus AWS, Google Cloud and Microsoft Azure. For prompts up to 100,000 tokens it costs $0.10 per million input tokens and $0.50 per million output tokens, which is 90 percent below Haiku 4.5, and 50 percent below for prompts over that threshold, which Anthropic averages to about 75 percent cheaper to run because the updated tokenizer uses slightly more tokens per task. Haiku 5.5 is the first Haiku-class model with an adjustable effort setting, the Python and TypeScript SDKs add computer use and browser use in beta, and Anthropic positions Sonnet 5.5 and Opus 5.5 as the better choices for complex agentic coding with Haiku 5.5 for compaction, summarization, classification and subagent work.
  • Why it matters: Anthropic separately halved Sonnet 5.5 cache reads from $0.20 to $0.10 per million tokens, which it states cuts the cost of most agentic work on that model by around 20 percent and which needs no migration, and it states Max and Team subscribers get a monthly API credit rolling out this week at $100 for Max 5x, $200 for Max 20x and up to $500 pooled for Team.
  • Follow-up: The open item is whether the vendor-run benchmark figures, whose method is described in the Haiku 5.5 system card, hold in independent testing.