Claude 4 (Opus 4 / Sonnet 4)
Hybrid reasoning and hours-long agentic coding sessions.
You're looking at an AI model record: who built it, what it can do, and the hardware and data behind it.
Claude 4 (May 2025) fused instant responses with extended thinking in one model and pushed agentic coding — Opus 4 could sustain multi-hour autonomous engineering tasks. Anthropic shipped it under its ASL-3 safety standard, the first activation of that deployment tier.
How this connects to the rest of the stack
This record's links at a glance — click a group to jump to it.
Developed by
Succeeded by
Successor to
Competes with
Sources
Record reviewed as of 2026-01-15 — see Methodology.