Anthropic/Model/Claude/4.5/Opus

• Claude Opus 4.5

  • Key Features

    • Released in November 2025
    • Extended Thinking capability: Allows the model to deliberate on complex problems before generating an output
    • Effort Parameter: Users can control reasoning depth with Low, Medium, and High effort settings, balancing speed and intelligence
    • Multilingual support: Handles multiple languages effectively
    • Vision processing: Processes and understands visual inputs
    • Enhanced reasoning: Excels in complex reasoning tasks
    • Context window: 200,000 tokens
    • Maximum output: 64,000 tokens
    • Knowledge cutoff: March 2025
    • Model version status: Stable release
    • Tool Search Capability: Enables dynamic discovery and utilization of tools without preloading all definitions, improving efficiency

  • Performance Highlights

    • Industry-leading performance on complex reasoning tasks
    • Enhanced Extended Thinking mode for improved response quality on complex tasks
    • Optimized for advanced coding, agentic search, and creative writing

  • Model Tiers

    • claude-opus-4-5
      • Full-featured flagship model
      • Best performance across all metrics
    • Previous versions:
      • Anthropic/Model/Claude/4/Opus
      • Anthropic/Model/Claude/Sonnet/4.5

  • Access Details

    • API availability: Available via claude-opus-4-5-20251101 in the Claude API
    • Pricing:
      • Input tokens: $5.00 per million tokens
      • Output tokens: $25.00 per million tokens
    • Available across multiple platforms:
      • Anthropic API: claude-opus-4-5-20251101
      • AWS Bedrock: anthropic.claude-opus-4-5-20251101-v1:0
      • Google Vertex AI: claude-opus-4-5@20251101
      • Azure AI Foundry: claude-opus-4-5-20251101
    • SDK support:
      • Python: pip install -U "anthropic"
      • TypeScript: npm install @anthropic-ai/foundry-sdk
    • Effort parameter options:
      • Low Effort: For simple tasks requiring quick responses
      • Medium Effort: Balances speed and reasoning depth
      • High Effort: Maximizes reasoning capabilities for complex tasks

  • Technical Specifications

    • Model architecture: Details not specified
    • Training data: Details not specified
    • Input/output formats: Text, images, code
    • Special capabilities:
      • Extended Thinking mode for complex problem-solving
      • Configurable effort levels (Low, Medium, High)
      • Dynamic tool discovery and utilization
      • Enhanced multilingual support
      • Vision processing capabilities
      • Advanced reasoning capabilities
    • Model version code: claude-opus-4-5-20251101
    • Safety measures: Operates under strict safety protocols, including advanced cybersecurity measures and enhanced detection algorithms for harmful queries, as part of Anthropic's Responsible Scaling Policy

  • References

    • Official documentation
    • Claude in Microsoft Foundry
    • Claude Opus 4.5 in Amazon Bedrock