Anthropic has launched Claude Opus 4.8, a direct challenger to OpenAI's GPT-5.5 and Google's Gemini 3.1 Pro. The new model focuses on agentic coding, large-scale automation, and improved reasoning.
According to a comparison chart published by Anthropic, Opus 4.8 leads most benchmark categories, but GPT-5.5 outperforms it in agentic terminal coding. However, Opus 4.8 excels in multidisciplinary reasoning scores and knowledge work tasks.
Anthropic emphasizes the model's 'honesty improvements,' stating that Opus 4.8 is 'around four times less likely' to allow flawed code to pass without flagging issues.
The company has also introduced customizable 'effort' settings, allowing users to determine how much computational power the model spends on a task. Additionally, a research-preview feature called 'dynamic workflows' enables the system to coordinate hundreds of parallel AI subagents within one coding session.
Pricing for Opus 4.8 remains the same as Opus 4.7, with fast mode now 'three times cheaper' operationally while delivering responses at 2.5× the speed.
