
Anthropic 发布 Claude Sonnet 5 后,官方系统卡给出了与 Sonnet 4.6、GPT-5.5、Gemini 3.5 Flash 在 SWE-bench Pro、Terminal-Bench、HLE、OSWorld 等 benchmark 上的可比结果。
claude-sonnet-5;发布期价格为每百万输入 token 2 美元、每百万输出 token 10 美元,2026 年 8 月 31 日后回到 3 美元 / 15 美元。1| Benchmark / 口径 | Claude Sonnet 5 | Claude Sonnet 4.6 | Claude Opus 4.8 | GPT-5.5 | Gemini 3.5 Flash | 其他模型 | 来源 |
|---|---|---|---|---|---|---|---|
| SWE-bench Pro | 63.2% | 58.1% | 69.2% | 58.6% | 55.1% | 12 | |
| SWE-bench Verified | 85.2% | 2 | |||||
| SWE-bench Multilingual | 78.3% | 2 | |||||
| SWE-bench Multimodal | 28.1% | 2 | |||||
| Terminal-Bench 2.1 | 80.4% | 67.0% | 82.7% | 83.4%(Codex CLI harness) | 76.2% | 12 | |
| CursorBench | 61.2% | 49.0% | 63.8% | 2 | |||
| FrontierCode v1 | 38.8% | 15.1% | 25.5% | 2 | |||
| ProgramBench(hidden test pass rate,区间) | 76–86% | 52–74% | 80–90% | Mythos 5:84–93% | 2 | ||
| BrowseComp(single agent) | 84.7% | 76.2% | 84.4% | 2 | |||
| BrowseComp(multi agent) | 86.6% | 2 | |||||
| BrowseComp(10M-token,max effort) | 84.7% | 2 | |||||
| Humanity's Last Exam(no tools) | 43.2% | 34.6% | 49.8% | 41.4% | 40.2% | 12 | |
| Humanity's Last Exam(with tools) | 57.4% | 46.8% | 57.9% | 52.2% | 12 | ||
| USAMO 2026 | 79.5% | 55.0% | 96.7% | Mythos 5:99.8% | 2 | ||
| ArxivMath(no tools) | 65.7% | 2 | |||||
| ArxivMath(with tools) | 72.2% | 2 | |||||
| OSWorld-Verified | 81.2% | 78.5% | 83.4% | 78.7% | 78.4% | 12 | |
| AutomationBench | 13.5% | 5.3% | 12.9% | 14.5% | 2 | ||
| GDPval-AA v2(Elo) | 1618 | 1395 | 1615 | 1509 | 1357 | 12 | |
| GDP.pdf(no tools,mean criteria pass rate) | 67.5% | 66.9% | 2 | ||||
| GDP.pdf(with tools,mean criteria pass rate) | 81.6% | 78.6% | 2 | ||||
| Real-World Finance v2(Elo) | 1219 | 低 219 Elo | 1222 | Fable 5 胜 Sonnet 5:69% | 2 | ||
| ChartMuseum(no tools) | 70.1% | 59.3% | 75.8% | 2 | |||
| ChartMuseum(with tools) | 86.7% | 80.9% | 89.7% | 2 | |||
| CharXiv Reasoning(no tools) | 77.0% | 71.6% | 80.5% | 2 | |||
| CharXiv Reasoning(with tools) | 88.3% | 85.3% | 89.9% | 2 | |||
| BenchCAD Vision2Code(no tools,voxel IoU) | 0.266 | 0.267 | 2 | ||||
| BenchCAD Vision2Code(with tools,voxel IoU) | 0.373 | 0.327 | 2 | ||||
| OfficeQA | 73.3% | 68.7% | 2 | ||||
| OfficeQA Pro | 59.4% | 53.4% | 2 | ||||
| HealthBench Professional | 57.8% | 44.2% | 51.8% | 2 | |||
| Legal Agent Benchmark(Full Public Set,全通过率) | ⚠️ 8.9(System Card 摘要表)/ 8.92%(System Card 细分表) | 8.0% | 2 | ||||
| Legal Agent Benchmark(Full Public Set,mean criterion-pass) | 88.26% | 88.48% | 2 | ||||
| Legal Agent Benchmark(Harvey held-out,全通过率) | ⚠️ 5.8%(System Card 摘要表有)/ 细分表无数据 | 5.4%(摘要表) | 2.1% | 0.8% | 2 | ||
| Legal Agent Benchmark(Harvey held-out,mean criterion-pass) | 91.2% | 2 | |||||
| Toolathlon(Pass@1) | 54.3% | 49.4% | 59.9% | Fable 5:61.7%;Mythos 5:61.7% | 2 | ||
| Toolathlon(Pass@3) | 63.0% | 60.2% | 67.6% | Fable 5:68.5%;Mythos 5:66.7% | 2 | ||
| Toolathlon(Pass³) | 40.7% | 38.0% | 48.1% | Fable 5:55.6%;Mythos 5:58.3% | 2 | ||
| Toolathlon(平均轮次,越少越好) | 26.0 | 16.5 | 24.5 | Fable 5:19.8;Mythos 5:19.0 | 2 | ||
| AA-Briefcase(Elo) | 1393 | 1352 | Fable 5:1586 | 2 |
anthropic.com
www-cdn.anthropic.com
This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.