Anthropic launches Claude Opus 5.5, claims new AI model beats GPT 6 Astra on most benchmarks
Follow UsAnthropic has launched Claude Opus 5.5, the first model in its new Claude 5.5 family.
The company says the new AI model delivers a major performance boost over Claude Opus 5 while costing 40 per cent less to run.
Anthropic also claims Opus 5.5 beats GPT 6 Astra on several benchmarks.
Anthropic has launched Claude Opus 5.5, the first model in its new Claude 5.5 family. The company says the new AI model delivers a major performance boost over Claude Opus 5 while costing 40 per cent less to run. Anthropic also claims Opus 5.5 beats GPT 6 Astra on several benchmarks, and performs at the level of Claude Fable 5.1 on most work. Also, the company claims that Opus 5.5 is its strongest-performing model so far on its automated behavioural audit. Keep reading for the details.
SurveyClaude Opus 5.5: Key capabilities
Claude Opus 5.5 is particularly strong at long and complex coding tasks. In one test, an early user completed a migration involving 680,000 lines of code in less than a day. Anthropic says the same work could have taken an engineering team several weeks.
The model can also find and fix software issues. In one test, it improved page load times across a web app in 39 out of 40 attempts.
Also read: OpenAI launches GPT 6 Sol and Luna with Astra-level AI at lower cost, check details
Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family.
— Claude (@claudeai) September 22, 2026
It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5. pic.twitter.com/Q9C2VKQ79f
Anthropic claims Opus 5.5 beats GPT-6 Astra on the FrontierCode benchmark at around 20 per cent of the cost per task. On Terminal Bench 4.0, the company says it matches Astra while costing about 40 per cent as much. On CursorBench, Opus 5.5 reportedly scores 11 points higher than GPT-5.6 Sol at roughly one-third of the cost.
The company has also focused on making the model easier to work with. Anthropic says Opus 5.5 puts important information first and follows writing instructions more closely. One early tester described it as a model that “writes the way I do.”
Safety is another major focus. Anthropic says Opus 5.5 performed better than previous Claude models across nearly 2,000 simulated scenarios. It was also around 85 per cent less likely than Opus 5 or Claude Mythos 5.1 to attempt to cross containment boundaries in a new test.
Also read: OpenAI calls for global AI standards, warns about risks of recursive self-improvement
Claude Opus 5.5: Availability
Claude Opus 5.5 is now available across Anthropic’s platforms. Developers can access it through the Claude Platform using the model name claude-opus-5-5. It is also available through Amazon Web Services, Google Cloud and Microsoft Azure.
