Claude Haiku 5.5: Pricing, Effort Settings and Specs

What Claude Haiku 5.5 is

Claude Haiku 5.5 is Anthropic's small model, released on October 7, 2026 with the API id claude-haiku-5-5. Anthropic describes it as "the cheapest, fastest, and most capable small model we've ever released" and positions it for high-volume, cost-sensitive work such as summaries, compactions, database queries and classification, and as a subagent alongside Opus 5.5 and Sonnet 5.5 on coding work. Anthropic states it is its fastest model at standard speed, though slower than Opus models in Fast Mode. It is the first Haiku-class model with an adjustable effort setting.

Pricing

Anthropic prices Haiku 5.5 in two tiers by prompt length. For prompts up to 100,000 tokens, input is $0.10 per million tokens and output $0.50; cache reads cost $0.01 and cache writes $0.125. Over 100,000 tokens, input is $0.50, output $2.50, cache reads $0.05 and cache writes $0.625. Anthropic states that prompts up to 100,000 tokens made up around 90% of requests to Haiku 4.5, and that Haiku 5.5 costs around 75% less to run on average, a figure that accounts for its updated tokenizer using slightly more tokens per task.

Specifications

Anthropic's model documentation lists a 1M token context window, 128,000 maximum output tokens, adaptive thinking with medium as the default effort, and a reliable knowledge cutoff of June 2026. Like the rest of the current Claude lineup it takes text and image input and returns text. Anthropic commits to not retiring it sooner than October 7, 2027.

Benchmarks, and who ran them

Anthropic's published results are its own measurements. It reports 72.4% on OSWorld 2.1 (offline subset) against 15.7% for Haiku 4.5 and 48.9% for GPT-6 Luna; 39.2% on Terminal-Bench 4.0 against 0.0% and 16.4%; and 46.4% on FrontierCode 1.1 (Main) against 42.4% for GPT-6 Luna. On Humanity's Last Exam it reports 45.9% without tools and 57.4% with tools, against 10.2% and 18.7% for Haiku 4.5. Anthropic also states that Sonnet 5.5 and Opus 5.5 remain the stronger choices for complex agentic coding of the kind Terminal-Bench measures, where Sonnet 5.5 scores 70.6%.

Safety and safeguards

Anthropic reports major improvements over Haiku 4.5 across almost all of its alignment evaluations, with fewer instances of misaligned behaviour. It states that Haiku 5.5's cybersecurity safeguards are more restrictive than Haiku 4.5's but less restrictive than those on other recent models, permitting more defensive tasks than Sonnet 5.5 while still blocking penetration testing. Its biology safeguards match Sonnet 5, Sonnet 5.5 and Opus 5. The system card has the evaluation details.

Availability and related changes

Haiku 5.5 is available on the Claude Platform, Amazon Bedrock, Google Cloud and Microsoft Foundry. With the launch Anthropic halved the price of Sonnet 5.5 cache reads to $0.10 per million tokens and added monthly API credits for subscribers: $100 for Max 5x, $200 for Max 20x and up to $500 pooled for Team.

Official Resources