Claude Opus 5.5 is the model Anthropic recommends for most work: long running agentic coding and knowledge work. Anthropic released it on September 22, 2026, and says it performs at the level of Claude Fable 5.1 on most work while costing 40 percent less to run than Opus 5.
Key facts
| Item | Claude Opus 5.5 |
|---|---|
| Maker | Anthropic |
| Released | September 22, 2026 |
| API model ID | claude-opus-5-5 |
| Context window | 1 million tokens |
| Max output | 128,000 tokens |
| Knowledge cutoff | June 2026 |
| Default effort | medium, with adaptive thinking always on |
| API price per million tokens | $4 input, $20 output, $0.20 cache reads |
Sources: Anthropic's models overview and pricing page, plus the Opus 5.5 announcement and the fast mode docs, checked October 24, 2026.
Pricing details
Standard pricing is $4 per million input tokens and $20 per million output tokens, with cache reads at 5 percent of the input price. Batch requests are half price. Opus 5.5 also has a generally available fast mode that runs the same model up to 2.5 times faster on output, for $8 per million input tokens and $40 per million output tokens. Fast mode works on the Claude API, Microsoft Foundry and the Anthropic managed Claude on Google Cloud, but not with the Batch API.
Where to use Claude Opus 5.5
- API and clouds: on the Anthropic API, Amazon Bedrock, Claude Platform on AWS, Google Cloud and Microsoft Foundry.
- Claude apps: included on Claude Pro ($20 a month), Max, Team and Enterprise. The Free plan offers Sonnet and Haiku instead.
- Claude Code: fast mode is toggled with
/fastand billed from usage credits.
How it compares
Opus 5.5 sits between Claude Fable 5.1 and Claude Sonnet 5.5. Against OpenAI, it is priced between GPT-6.1 Sol and GPT-6 Astra. For details, read Claude Fable vs Opus vs Mythos, which also answers "Mythos vs Opus 5.5", and see the release timeline.
Who should use it
Start with Opus 5.5 for coding, analysis and writing. Move to Fable 5.1 only when Opus falls short in your own tests at higher effort.