Claude Haiku 5.5: price, features and how it is used
Claude Haiku 5.5 is Anthropic’s fastest model, designed for high-volume, low-latency tasks such as classification, data extraction and routing. Released on October 7, 2026.
Provider: Anthropic · Released: October 2026
Results by test
| Test | Result | Source |
|---|---|---|
| OSWorld 2.1 (offline subset) | 72,4% | Anthropic, Claude Haiku 5.5 announcement |
| Humanity's Last Exam (no tools) | 45,9% | Anthropic, Claude Haiku 5.5 announcement |
| Humanity's Last Exam (with tools) | 57,4% | Anthropic, Claude Haiku 5.5 announcement |
| Terminal-Bench 4.0 | 39,2% | Anthropic, Claude Haiku 5.5 announcement |
| FrontierCode 1.1 (Main) | 46,4% | Anthropic, Claude Haiku 5.5 announcement |
| Chartography (no tools) | 46,4% | Anthropic, Claude Haiku 5.5 announcement |
Intelligence, price and speed by effort level
There is no data yet for this model on effort level, intelligence and price per task.
Data on the Intelligence Index, capability indexes, price per task, speed and latency: Artificial Analysis. Prices per 1M tokens are from official vendor data.
Compare Claude Haiku 5.5 with another model
Choose a model and see prices, context and results side by side.
How to use Claude Haiku 5.5
Claude Haiku 5.5 is available via the Claude Platform (API) and via AWS, Google Cloud and Microsoft Azure. To get started, create an API key in the Claude Platform.
- Create an account and an API key in the Claude Platform.
- In your Messages API request, set the model to claude-haiku-5-5.
- Effort is set via output_config.effort with levels max, xhigh, high, medium and low. The default is medium.
- The documentation recommends low for chat and simple high-volume requests, and high for longer agentic tasks and strict instruction following.
- Thinking can be turned off with thinking type disabled at high level or lower. At xhigh and max this returns a 400 error.
- For non-English text, keep in mind that non-English text uses more tokens than English, so the same text costs more. Check the quality of answers on your own real examples.
What it is suited for
- Lowest price in Anthropic’s lineup.
- Marked by Anthropic as the fastest model.
- Context window of 1M tokens.
What to keep in mind
- The price is for requests with up to 100,000 input tokens. Above this threshold the input is $0.50, and the output $2.50 per 1M tokens.
- It uses the newer tokenizer and the same text counts as about 30% more tokens compared with Haiku 4.5.
- Default effort level: medium. Anthropic shows per-level results in charts, but not as numbers.
- The page does not say at which effort level the results in the table were obtained.
Frequently asked questions about Claude Haiku 5.5
- How much does Claude Haiku 5.5 cost?
- For requests up to 100K tokens it is 0.10 USD per 1M input and 0.50 USD per 1M output. Above 100K tokens the price is 0.50 and 2.50 USD. Cache reads cost 0.01 USD.
- Is Claude Haiku 5.5 free?
- Not through the API, which is paid per token. The official page does not describe availability in the apps or a free plan, so I cannot confirm it.
- How do you use Claude Haiku 5.5?
- Write in your own language and check the quality on real examples. Tokens in many languages are more numerous than in English, but at these prices the cost stays low.
- How does Claude Haiku 5.5 differ from Claude Sonnet 5.5?
- Haiku is much cheaper: 0.10 and 0.50 dollars versus 2 and 10 for Sonnet. Haiku allows thinking to be turned off at high and lower, while Sonnet does not support disabled. Both have a 1 million token context.
More models in AI Benchmark
- Claude Fable 5.1 (Anthropic)
- Claude Opus 5 (Anthropic)
- Claude 4.5 Haiku (Anthropic)
- Claude Fable 5.2 (Anthropic)
- Claude Opus 5.5 (Anthropic)
- Claude Sonnet 5.5 (Anthropic)
Intelligence Index, token, TTFT and effort are explained in the AI Benchmark glossary. Prices are per 1M tokens and may change, so check the provider’s official page.
Sources: Anthropic, model overview · Anthropic, pricing · Anthropic, Haiku 5.5 page · Anthropic, Haiku 5.5 announcement
Last checked: 18.11.2026
Compare with other models
Set Claude Haiku 5.5: price, features and how it is used to another model by price, context and benchmarks.