Anthropic logo

Claude Opus 4.6

AnthropicProprietaryPending Human Review

Claude Opus 4.6 is the independently released Opus update for sustained coding, reasoning, and professional agent tasks. It improves planning, code review, debugging, and work in large repositories. Adaptive thinking and effort settings control computation, while context compaction supports longer workflows. The million-token context launched in beta and is now listed as standard in current API documentation, with 128K normal output and an optional 300K batch beta. The model accepts text and images and generates text.

2026-02-05
Unreleased
Undisclosed proprietary architecture
Proprietary

Specifications

Parameters
Unreleased
Architecture
Undisclosed proprietary architecture
License
Proprietary
Context Window
1,000,000 tokens
Max Output
128,000 tokens
Training Data Cutoff
Aug 2025
Type
multimodal
Modalities
textimage

Benchmark Scores

Advanced Specifications

Model Family
Claude
API Access
Available
Chat Interface
Available
Multilingual Support
Yes
Variants
claude-opus-4-6

Capabilities & Limitations

Capabilities
deep reasoningagentic codingcomplex codingresearchadaptive thinkingeffort controlcontext compactioncomputer usetool useparallel tool executionextended thinkingvisionmultilinguallong context
Known Limitations
higher latency than Sonnet and Haikumay overthink simpler tasks at default high effortknowledge cutoff limits real-time data without tools
Notable Use Cases
agentic codingdeep researchcomplex system architecturemulti-step agentic workflowsfinancial analysiscybersecuritylong-context document analysiscode review and debugging
Function Calling Support
Yes
Tool Use Support
Yes

Related Models