Claude Opus 5 by Anthropic: complex agentic coding with adjustable effort settings, near-frontier intelligence, and software engineering performance.
Capabilities, design details, and architectural traits
Claude Opus 5 is positioned as the default model for complex agentic coding and enterprise work, delivering near-frontier intelligence. It replaces Opus 4.8 with gains across agentic coding, computer use, long-horizon knowledge work, and mathematical and scientific reasoning.
| Trait | Detail |
|---|---|
| Adjustable effort settings | Customers can tune effort level (high, xhigh, max) to optimize for intelligence or conserve tokens for faster results |
| Source-code vulnerability discovery | Permitted at all access levels, while vulnerability discovery in compiled binaries remains blocked |
| Self-verifying agency | Demonstrated writing its own computer vision pipeline to extract geometry from raw pixels when denied a direct view of a drawing |
The effort setting is central to Opus 5's design.
Its safeguard configuration is distinct: it allows source-code vulnerability discovery to support defensive security work while blocking discovery in compiled binaries, which is more commonly used offensively.
Independent evaluations · Artificial Analysis
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Anthropic | Anthropic | OpenAI |
| Release Date | July 24, 2026 | June 9, 2026 | July 9, 2026 |
| Knowledge Cutoff | May 2026 | - | Feb 2026 |
| Context & Limits | |||
| Context Window | 1M | 1M | 1.1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | $5 Best Input Pricing | $10 | $5 Best Input Pricing |
| Output Pricing | $25 Best Output Pricing | $50 | $30 |
| Modalities | |||
| Inputs | textimage | textimagefile | textimage |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 63.1 Best Intelligence Index | 62.1 | 60.9 |
| Coding Index | 78.0 Best Coding Index | 76.5 | 77.4 |
| Agentic Index | 59.2 Best Agentic Index | 56.6 | 57.8 |
Graduate-level reasoning and expert Q&A evaluation.
Extremely difficult logical reasoning and knowledge.
Logical reasoning over long context windows.
Benchmark scores are independent evaluations sourced from Artificial Analysis. Intelligence, Coding, and Agentic indices reflect composite performance ratings (0–100).
