Claude Opus 4.7
Anthropic's most capable commercial model with frontier coding, vision, and long-horizon agentic performance
Claude Opus 4.7
Anthropic • April 2026
Training Data
Up to early 2026
Claude Opus 4.7
April 2026
Parameters
~750 billion
Training Method
Advanced Constitutional AI
Context Window
1,000,000 tokens (beta)
Knowledge Cutoff
Not disclosed
Key Features
Xhigh Effort Level • Task Budgets • Ultrareview for Claude Code • Enhanced Vision (3.75MP)
Capabilities
Reasoning: Maximum
Coding: Outstanding (+13% SWE-bench vs 4.6)
Vision: Outstanding (2576px)
Agentic: Long-horizon autonomy
What's New in This Version
+13% on SWE-bench Verified vs Opus 4.6, 3× production tasks on Rakuten-SWE, CursorBench 70% (up from 58%), 3× higher image resolution
Anthropic's most capable commercial model with frontier coding, vision, and long-horizon agentic performance
What's New in This Version
+13% on SWE-bench Verified vs Opus 4.6, 3× production tasks on Rakuten-SWE, CursorBench 70% (up from 58%), 3× higher image resolution
Technical Specifications
Key Features
Capabilities
Other Anthropic Models
Explore more models from Anthropic
Claude 3 Opus
Anthropic's most capable model with exceptional reasoning abilities
Claude 3 Sonnet
Anthropic's balanced model offering strong performance and speed
Claude 3 Haiku
Anthropic's fastest model optimized for speed and efficiency
