Flagship Claude model for deep reasoning, coding, and long-horizon agents
Benchmarks
| Benchmark | Category | Score | Variant | Provenance | Source | Observed |
|---|---|---|---|---|---|---|
| AA Intelligence Index | intelligence | 29.1 | thinking | independent | aa | Sep 9, 2026 |
| 23.7 | — | independent | aa | Sep 9, 2026 | ||
| Design Arena — 3d | preference | 1242 | — | mirror | openrouter | Sep 13, 2026 |
| Design Arena — androidnative | preference | 1190 | — | mirror | openrouter | Sep 19, 2026 |
| Design Arena — asciiart | preference | 1208 | — | mirror | openrouter | Sep 20, 2026 |
| Design Arena — codecategories | preference | 1256 | — | mirror | openrouter | Sep 21, 2026 |
| Design Arena — dataviz | preference | 1258 | — | mirror | openrouter | Sep 16, 2026 |
| Design Arena — fullstack | preference | 1166 | — | mirror | openrouter | Sep 15, 2026 |
| Design Arena — gamedev | preference | 1252 | — | mirror | openrouter | Sep 19, 2026 |
| Design Arena — mobileapps | preference | 1218 | — | mirror | openrouter | Sep 19, 2026 |
| Design Arena — svg | preference | 1203 | — | mirror | openrouter | Sep 21, 2026 |
| Design Arena — uicomponent | preference | 1254 | — | mirror | openrouter | Sep 20, 2026 |
| Design Arena — webapps | preference | 1174 | — | mirror | openrouter | Sep 16, 2026 |
| Design Arena — website | preference | 1259 | — | mirror | openrouter | Sep 11, 2026 |
| LMArena Text | preference | 1447 | thinking-32k | crowd | lmarena | Aug 10, 2026 |
| 1450 | — | crowd | lmarena | Sep 13, 2026 | ||
| LMArena WebDev | preference | 1495 | thinking-32k | crowd | lmarena | Aug 10, 2026 |
| 1468 | — | crowd | lmarena | Sep 11, 2026 | ||
| Median output speed | speed | 51 tok/s | thinking | independent | aa | Sep 21, 2026 |
| 47 tok/s | — | independent | aa | Sep 21, 2026 | ||
| Median time to first token | speed | 1.21s | — | independent | aa | Sep 20, 2026 |
| 15.87s | thinking | independent | aa | Sep 21, 2026 |
Includes index data from Artificial Analysis.
Providers
Live operational stats per hosting endpoint, via OpenRouter.
| Provider | Quant | Context | $/1M in · out | Uptime 24h | Latency | Throughput |
|---|---|---|---|---|---|---|
| Amazon Bedrock | — | 200K | $5 · $25 | 99.95% | — | — |
| Amazon Bedrock | — | 200K | $5.5 · $27.5 | — | — | — |
| Anthropic | — | 200K | $5 · $25 | 99.67% | — | — |
| Azure | — | 200K | $5 · $25 | — | — | — |
| Claude Platform on AWS | — | 200K | $5 · $25 | 98.84% | — | — |
| — | 200K | $5 · $25 | 99.89% | — | — |