Flagship quality for everyday work
We have added Claude Opus 5.5, the first model of Anthropic's new Claude 5.5 family, to Ayeto. Anthropic says it performs at the level of Claude Fable 5.1 on most work, costs about 40% less to run than Opus 5 and generates output more than 30% faster.
In practice, that means a model that can take on long coding tasks, research and document work that previously called for Fable, at a price much closer to everyday use.
What is Claude Opus 5.5?
Claude Opus 5.5 is Anthropic's flagship model for complex knowledge work, agentic coding and long-running autonomous tasks. It accepts text and images, works with tools and supports reasoning effort from low to max.
Anthropic also points to clearer communication: Opus 5.5 puts the most important information at the start of an answer, so longer responses are easier to scan.
How it compares
| Benchmark | Opus 5.5 | Fable 5.1 | Opus 5 |
|---|---|---|---|
| Terminal-Bench 4.0 | 66.4% | 55.8% | 52.3% |
| CursorBench 4.0 | 57.8% | 51.8% | 46.6% |
| FrontierCode v1.1 | 54.4% | 50.3% | 48.0% |
| GDPval-AA v2.1 (Elo) | 1846 | 1735 | 1708 |
| AutomationBench | 40.0% | 31.4% | 26.9% |
| Humanity's Last Exam (with tools) | 67.7% | 65.6% | 63.6% |
Coding and terminal work
On Terminal-Bench 4.0, Opus 5.5 reaches 66.4%, ahead of both Fable 5.1 and Opus 5. According to Anthropic, at medium effort it matches what Opus 5 achieved at maximum effort, so many tasks finish with fewer tokens.
Knowledge work
On GDPval-AA, which measures professional tasks such as analyses, reports and presentations, Opus 5.5 leads with 1846 Elo. It is a strong fit for research, drafting documents and working through large sets of sources.
Safety
In Anthropic's automated behavioural audit, Opus 5.5 is the best-performing model the company has tested. It tried to get around the limits of its environment 85% less often than Opus 5 and matches Fable 5.1 in resistance to prompt injection, which matters whenever an assistant reads web pages, e-mails or documents.
Two variants in Ayeto
In Ayeto you will find Claude Opus 5.5 and Claude Opus 5.5 (High). The standard variant is the right default for most work. The High variant spends more reasoning on each request and suits the hardest problems: complex debugging, larger refactors or analyses where precision matters more than speed.
When to choose Opus 5.5 in Ayeto
Choose Opus 5.5 for long agentic tasks, work with code and repositories, research across many sources and documents that must be accurate and well structured. If you have been using Fable 5.1 for this kind of work, Opus 5.5 is worth testing as a less expensive alternative.
For short answers, simple rewriting or classification, a lighter model will be more economical. In Ayeto, you can switch models in the assistant settings without changing the rest of its setup.
How its cost compares
Per token, Opus 5.5 is 20% cheaper than Opus 5, and reading from the prompt cache is 60% cheaper. Because it also needs fewer tokens to finish a task, Anthropic estimates the overall cost of a typical workload at about 40% below Opus 5.
Ayeto converts usage into credits. The most reliable way to decide is to run the same real task through Opus 5.5, Opus 5 and Fable 5.1 in Ayeto Model Comparison and compare quality, processing time and total credit usage.
Try Claude Opus 5.5
Claude Opus 5.5 is now available in Ayeto. Select it in your assistant settings or directly in the chat and give it the task you would normally hand to your strongest model.