Jump to a page, find an agent or model, or run an action
Anthropic's deepest reasoning model.
Opus 4.8 is Anthropic's most capable model — built for the hardest reasoning, the longest documents, and the code changes you don't want to review twice. It's slower and pricier than Sonnet, and worth it when correctness is the whole point.
A 500K context window means it can hold an entire mid-size codebase or a full research corpus in a single call.
Claude Opus 4.8 shows its cost before every call and covers you with a hard cap — or call free under an active time pass.