Claude Sonnet 5 by Anthropic: most agentic Sonnet, nearing Opus 4.8. Plans, uses tools, self-checks output, finishes tasks prior Sonnets abandoned.
Capabilities, design details, and architectural traits
Claude Sonnet 5 is built to be the most agentic Sonnet model yet, narrowing the gap between Sonnet and Opus tiers. It can make plans, use tools like browsers and terminals, and run autonomously at a level that previously required larger models. Its performance approaches that of Opus 4.8.
| Trait | Detail |
|---|---|
| Agentic autonomy | Finishes complex tasks where previous Sonnet models would stop short; completes end-to-end workflows that previously stalled halfway |
| Self-checking output | Checks its own output without explicitly being asked |
| Sustained coding and debugging | Handles multi-step software engineering work across messy technical contexts with strong follow-through |
| Reduced cybersecurity capability | Much lower ability to perform cybersecurity tasks than current Opus models, by design |
| Lower undesirable behaviors | Overall lower rate of undesirable behaviors than Sonnet 4.6, safer in agentic contexts |
Independent evaluations · Artificial Analysis
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Anthropic | Anthropic | Anthropic |
| Release Date | June 30, 2026 | July 24, 2026 | June 9, 2026 |
| Knowledge Cutoff | - | May 2026 | - |
| Context & Limits | |||
| Context Window | 1M | 1M | 1M |
| Pricing (per 1M tokens) | |||
| Input Pricing | $2 Best Input Pricing | $5 | $10 |
| Output Pricing | $10 Best Output Pricing | $25 | $50 |
| Modalities | |||
| Inputs | text | textimage | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 55.3 | 63.1 Best Intelligence Index | 62.1 |
| Coding Index | 71.5 | 78.0 Best Coding Index | 76.5 |
| Agentic Index | 49.7 | 59.2 Best Agentic Index | 56.6 |
Graduate-level reasoning and expert Q&A evaluation.
Extremely difficult logical reasoning and knowledge.
Logical reasoning over long context windows.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
