System Card: Claude Opus 4.6

2026

Open publication workspace · Sign in to read the full PDF.

AI-generated summary

This system card details the capabilities and safety evaluations of Claude Opus 4.6, a frontier large language model.

* Comprehensive assessment of model capabilities across various domains.
* In-depth analysis of safety evaluations, including safeguards, user wellbeing, and alignment.
* Detailed exploration of specific capabilities such as software engineering, agentic tasks, and long context reasoning.

Tags: Claude Opus, AI safety, LLM evaluation, capabilities, alignment

Check the original publication for accuracy and context.