Anthropic Claude model for long-running agentic coding and knowledge work.
Claude Mythos 5
Limited-availability autonomous intelligence model delivering breakthrough problem-solving across scientific and technical disciplines.
Technical Specifications
Benchmark Evaluations
Deep Architectural Overview
Claude Mythos 5 represents Anthropic’s most advanced autonomous reasoning engine. Available to approved enterprise partners and Project Glasswing participants under specialized governance protocols.
Strengths & Considerations
- Frontier autonomous reasoning scores (87.4% GPQA Diamond, 85.6% SWE-bench)
- 1M token context with 128k output buffer
- Unparalleled scientific and technical hypothesis validation
- Limited availability to approved enterprise security programs
Token & API Pricing
Similar & Alternative Models
Explore other frontier models from Anthropic and comparable reasoning engines.
The ultimate model for demanding reasoning and long-horizon agentic work, featuring 0.025x cache reads ($0.25/MTok).
The most capable autonomous intelligence system engineered by Anthropic, restricted to Project Glasswing participants.
The premier enterprise flagship model for multi-hour autonomous coding, complex system engineering, and vision-heavy agent tasks.