Claude Opus 5.5
A next-generation flagship large language model in Anthropic's Opus tier within the Claude model family, designed for high-complexity reasoning and creative tasks. It serves as the successor iteration to Claude Opus 4.
Core Facts
Artificial Analysis gave Argon an intelligence index score of 53, on par with GPT-6 Astra.
Opus 5.5 reaches the level of Fable 5.1 on most tasks, but with approximately 40% lower operating costs and faster speed.
On Terminal-Bench 4.0, Claude Opus 5.5 leads with 66.4%, followed by GPT-6 Astra at 61.8% and Argon at 57.4%.
Timeline (last 90 days)
On the Artificial Analysis Intelligence Index (v5.0), Claude Opus 5.5 and GPT-6 Astra rank immediately behind Gemini 4 Carbon.
Claude Opus 5.5 and GPT-6 Astra rank immediately behind Gemini 4 Carbon on the leaderboard, indicating that the gap between leading models continues to narrow.
First-version Neural Yap benchmark results show that GPT-series models have the least filler (lowest ranking), while Claude Opus 5.5 is the model with the most filler.
In Willison's tests, Claude Opus 5.5, when asked to "Make it a LOT more fun," did produce a game he genuinely enjoyed playing.
Willison's initial impression is that GPT-6 Astra is not as good as Opus 5.5 at grasping "fun," and he expects OpenAI to catch up by the end of the year.
Battle Drones is a Command & Conquer-style real-time strategy game; the early version was built with Claude Fable 5, and after Opus 5.5 took over, the gameplay experience changed qualitatively.
PuraMetric is closer to SolidWorks; according to developer MaleficentRelease61 on Reddit, it is written in Rust and JavaScript, took only about one month to develop, and used Claude Opus 5.5 to assist development.
The original Reddit post did not provide proofs or paper links, so the stated content is an assertion rather than a verifiable academic result.
A post appeared in a Reddit community in which a user quoted Claude Opus 5.5's output saying, "If these results hold up, this will be the most important single day in the history of mathematics and theoretical computer science."
The prompt required no use of ready-made skills, no use of animation frameworks such as Remotion, no use of templates, and that all 3D objects be built from scratch.
40 more timeline events
All Facts (20)
Artificial Analysis gave Argon an intelligence index score of 53, on par with GPT-6 Astra.
80%VerifiedOpus 5.5 reaches the level of Fable 5.1 on most tasks, but with approximately 40% lower operating costs and faster speed.
80%VerifiedOn Terminal-Bench 4.0, Claude Opus 5.5 leads with 66.4%, followed by GPT-6 Astra at 61.8% and Argon at 57.4%.
75%VerifiedOpus 5.5 scored 66 on the Terminal agent benchmark, higher than Sonnet's 55 and Astra's 57.9.
70%VerifiedAnthropic states that Opus 5.5 can reduce total costs for typical tasks by approximately 40% and improve output speed by more than 30%.
70%VerifiedClaude Opus 5.5 generates output more than 30% faster than Opus 5.
70%VerifiedGPT-6 Astra's context window is approximately 1 million tokens.
70%VerifiedClaude Duet is powered by Claude Opus 5.5.
65%VerifiedGemini Agent allows users to choose the underlying model, including Gemini 4 Argon or Claude Opus 5.5
65%VerifiedHaiku 5.5 can work as a sub-agent alongside Claude Opus 5.5 and Sonnet 5.5 to complete programming tasks.
65%VerifiedIn a chart published by Google, Argon's Deep SWE score is close to 78%, while Claude Opus 5.5 and GPT-6 Astra are only slightly above 74%.
65%VerifiedIn the Automation Bench office workflow test, Gemini 4 Argon scored 51.3%, GPT-6 Astra scored 41.4%, and Claude Opus 5.5 scored 42.5%.
65%VerifiedWhen heavy users' consumption far exceeds their paid quota, every additional active user retained by the platform represents an additional subsidy burden.
65%VerifiedOpus 5.5's release page directly uses GPT-6's Astra as its comparison target.
65%VerifiedGPT-6 Luna can handle almost all tasks previously done with GPT-5.6 Soul, including making PPTs, video editing with subtitles, creating covers, automated publishing workflows, complex Skill modifications, simple project development, and researching new technologies.
65%VerifiedOn Terminal Bench 4.0, Opus 5.5 scored 66.4%, Astra scored 57.9%, Fable 5.1 scored 55.8%, and the previous generation Opus 5 scored 52.3%.
65%VerifiedFor performance range figures with extremely wide spans, reaching the upper end of the range often depends on favorable conditions such as specific tasks, specific baseline comparisons, and specific batch sizes.
65%VerifiedV4.1-Flash scores 1 point higher than a GPT-5.6-level model in comparison.
65%VerifiedOpus 5 is the flagship model of Anthropic's Claude series, known for its longer context window (capable of handling codebases of hundreds of thousands of tokens) and more cautious output style.
65%VerifiedNegative Prompting, by explicitly telling the model what not to do through exclusionary constraints, is often more precise and controllable than positive style descriptions.
65%