Claude Opus 5
claude-opus-5
Anthropic · reasoning · api
Intelligence
Context
1M
Max output
128K
Input $/1M
$5.00
Output $/1M
$25.00
Modalities
text · image · document
Released
24 Jul 2026
Intelligence Index via Artificial Analysis · 0–100, higher is better
AI summary
● machine-written
Claude Opus 5: Anthropic's strongest Opus model for coding and agents
Claude Opus 5 is Anthropic's premium reasoning model designed for complex coding tasks, agentic workflows, and enterprise knowledge work. It features a 1M token context window and 128K token max output. Available on Claude Platform, AWS, Google Cloud, and Microsoft Foundry, it represents a significant capability upgrade within the Opus family.
What's new
- Stronger coding performance with production-ready code delivery
- Improved agentic capabilities with longer autonomous task execution
- Effort settings allow users to control compute allocation per task
- Available across Claude Platform, AWS, Google Cloud, and Microsoft Foundry
Best for
Advanced coding and software engineeringProduction agentic workflows and multi-tool orchestrationLong-running enterprise tasks and knowledge workFinancial and genomics analysis
Common questions about Claude Opus →
Source: https://docs.claude.com/en/docs/about-claude/models/overview