Claude Sonnet 5.5: price, features and how it is used
Claude Sonnet 5.5 is Anthropic’s model, described as the best combination of speed and intelligence. Released on 28 September 2026. Supports text and images as input and text as output.
Provider: Anthropic · Released: September 2026
Results by test
| Test | Result | Source |
|---|---|---|
| Terminal-Bench 4.0 | 70,6% | Anthropic, announcement of Claude Sonnet 5.5 |
| FrontierCode 1.1 (Main, xhigh effort) | 52,1% | Anthropic, announcement of Claude Sonnet 5.5 |
| CursorBench 4.0 | 55,5% | Anthropic, announcement of Claude Sonnet 5.5 |
| Humanity's Last Exam (with tools) | 64,5% | Anthropic, announcement of Claude Sonnet 5.5 |
| OSWorld 2.1 (partial) | 80,1% | Anthropic, announcement of Claude Sonnet 5.5 |
| Chartography (no tools) | 61,6% | Anthropic, announcement of Claude Sonnet 5.5 |
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 Sonnet 5.5 with another model
Choose a model and see prices, context and results side by side.
How to use Claude Sonnet 5.5
Claude Sonnet 5.5 is available through the Claude Platform (API), in the Claude apps and Claude Code, and through AWS, Google Cloud and Microsoft Azure. For API use you need a key from the Claude Platform.
- Create an API key in the Claude Platform or open the Claude app.
- In the request, set the model to claude-sonnet-5-5.
- Effort is set with output_config.effort: max, xhigh, high, medium or low. The default is high in the API and medium in the apps and Claude Code.
- The documentation recommends high for harder tasks, medium or low for chat, and medium for well-described agentic coding tasks.
- Thinking cannot be turned off. Thinking mode with type between_tools works at low, medium and high, while at xhigh and max it returns error 400.
- 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
- Price $2 and $10 per 1M tokens with a 1M-token context window.
- Marked by Anthropic as a fast model.
- In the published table, its Terminal-Bench 4.0 result is higher than that of Opus 5.5.
What to keep in mind
- Default effort level: high. Per-level token, price and time figures are not published.
- On FrontierCode, Anthropic reports a lower score at max tier (46.2%) than at xhigh (52.1%).
- Setting temperature, top_p or top_k returns a 400 error.
Frequently asked questions about Claude Sonnet 5.5
- How much does Claude Sonnet 5.5 cost?
- USD 2 per 1M input and USD 10 per 1M output tokens. Cache read is USD 0.10 and cache write is USD 2.50.
- Is Claude Sonnet 5.5 free?
- Not through the API, which is paid per token. The official page says nothing about a free plan in the apps, so check in your own account what it includes.
- How do you use Claude Sonnet 5.5 in your own language?
- Write in your language and compare the result with what you expect. Tokens in non-English languages are more numerous, so the same text costs more than in English. Try low and medium levels for chat if you want to save.
- How does Claude Sonnet 5.5 differ from Claude Haiku 5.5?
- Both have a 1 million token context and up to 128 thousand output. Sonnet costs 2 and 10 dollars, and Haiku 0.10 and 0.50 for requests up to 100 thousand tokens. The default effort is high for Sonnet and medium for Haiku.
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 Haiku 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, Sonnet 5.5 page · Anthropic, announcement of Sonnet 5.5
Last checked: 18.11.2026
Compare with other models
Set Claude Sonnet 5.5: price, features and how it is used to another model by price, context and benchmarks.