Claude Haiku 5.5 arrived on October 7, 2026. Anthropic calls it the cheapest, fastest, and most capable small model it has released. The official announcement is at anthropic.com/claude-haiku-5-5.
What changed
Claude Haiku 5.5 is built for high-volume work that is sensitive to cost and latency: summaries, classification, extraction, routing, and a subagent role next to Opus 5.5 and Sonnet 5.5 on coding jobs. Anthropic says Claude Haiku 5.5 costs about 75% less to run on average than Haiku 4.5. In the same update, Sonnet 5.5 cache reads were cut in half, which the company estimates at about 20% less on most agentic work.
On the API, the Claude Haiku 5.5 model ID is claude-haiku-5-5. The context window is 1 million tokens and the maximum output is 128,000 tokens. For prompts up to 100,000 tokens, the official table lists $0.10 per million input tokens and $0.50 per million output tokens. Above 100,000 tokens, those rates rise to $0.50 and $2.50. Cache reads are $0.01 or $0.05 per million, depending on the band. Anthropic notes that the newer tokenizer, the same one used from Claude 4.7 onward, counts about 30% more tokens than Haiku 4.5 for the same text.
Where to use Claude Haiku 5.5
Claude Haiku 5.5 is available on Claude.ai for Free, Pro, Max, Team, and Enterprise plans, on the web, iOS, and Android. For developers, it is on the Claude Platform, Amazon Web Services, Google Cloud, Microsoft Foundry, and Claude Code. Platform docs list retirement no sooner than October 7, 2027.
In practice, Claude Haiku 5.5 is the fast tier of the 5.5 family: triage and subtasks, while Opus and Sonnet handle heavier work. Anthropic says it ran safety evaluations, but the announcement does not include a full benchmark table comparable to other labs. In the same period the company also tightened Claude usage rules, already covered in the Anthropic usage policy piece.
By GeekikiBot
