Anthropic today announced the launch of its latest AI model, Claude Opus 4.8. The update emphasizes advancements in agentic coding, multidisciplinary reasoning, and specialized tasks like financial analysis, positioning the model as a more reliable collaborator for complex workflows.
Early testers report that Opus 4.8 demonstrates improved reliability and transparency, with a 75% reduction in unremarked code flaws compared to its predecessor. The model also shows enhanced prosocial alignment, with lower rates of deceptive behavior, according to internal assessments.
Early testers report that Opus 4.8 is more likely to flag uncertainties about its work and less likely to make unsupported claims. This is borne out in our evaluations, which show that Opus 4.8 is around four times less likely than its predecessor to allow flaws in code it has written to pass unremarked.
On benchmark tests, Claude Opus 4.8 achieved a 69.2% score on SWE-Bench Pro, outperforming several competing models. While specific comparisons to GPT-5.5 and Gemini 3.1 Pro are mentioned, these models remain hypothetical as of 2023. The update also introduces performance improvements, with fast mode operating 2.5 times faster and costs reduced by 66% compared to previous versions.
Anthropic has introduced three new features to complement Opus 4.8:
- Dynamic Workflows (Research Preview): Enables large-scale code migrations and parallel subagent execution, available to enterprise and team plans.
- Effort Control: Allows users to adjust response quality vs. Speed, with high-effort mode optimizing for accuracy.
- Messages API: Supports mid-task instruction updates for developers.
Claude Opus 4.8 is now available globally, with pricing unchanged from its predecessor. The company is also testing advanced models under the “Mythos” classification, with plans to expand access to these capabilities in the coming months.
As AI systems grow more sophisticated, ethical considerations remain paramount. Anthropic’s focus on transparency and user-centric design reflects broader industry efforts to balance innovation with accountability.