Anthropic released Claude Haiku 5.5. It cut API prices by 90% for requests below 100,000 tokens. Input tokens now cost $0.10 per million and output tokens $0.50 per million. These rates match OpenAI’s GPT-6 Luna. Above 100,000 tokens, prices drop by 50% instead. Sonnet 5.5 cache reads also got cheaper. Max and Team subscribers now receive monthly API credits. Haiku 5.5 launches on Anthropic’s platform, AWS, Google Cloud and Azure. It blocks penetration testing under standard safeguards. Benchmarks show gains over Haiku 4.5 but Sonnet 5.5 remains ahead. Customers report faster results and lower latency. Anthropic positions Haiku as a support model for enterprise agents.
What changed
Haiku 4.5 cost $1.00 per million input tokens and $5.00 per million output tokens for all requests.
What it unlocks
Running repetitive, short-context AI tasks at far lower cost.
- 90% API price reduction for requests below 100,000 tokens
- $0.10 per million input tokens, $0.50 per million output tokens
- 75% lower average workload cost vs Haiku 4.5
- 90% of Haiku 4.5 requests fall below 100,000 tokens
What you need to act on it
- API access via Anthropic, AWS, Google Cloud, or Microsoft Azure
Sources