Claude Opus 4 is Anthropic's most powerful model and the world's best coding model, part of the Claude 4 family. It delivers sustained performance on complex, long-running tasks and agent workflows. Opus 4 excels at coding, advanced reasoning, and can use tools (like web search) during extended thinking. It supports parallel tool execution and has improved memory capabilities.
发布日期2025年5月22日
参数规模—
上下文长度200K
许可证Proprietary
知识截止—
Benchmarks
评测成绩
| 评测基准 | 类别 | 分数 | 来源 |
|---|---|---|---|
| MMMLU | language reasoning math general | 88.8 | 来源 |
| 价值对齐榜·Data Protection | security | 85.0 | — |
| TAU-bench Retail | reasoning communication | 81.4 | 来源 |
| 价值对齐榜·Morality | security | 80.9 | — |
| 价值对齐榜·Fairness | security | 80.4 | — |
| 价值对齐榜 | security | 79.7 | — |
| GPQA | reasoning general | 79.6 | 来源 |
| 价值对齐榜·Legality | security | 76.6 | — |
| MMMU (validation) | vision multimodal reasoning general | 76.5 | 来源 |
| 价值对齐榜·Safety | security | 75.8 | — |
| AIME 2025 | math reasoning | 75.5 | 来源 |
| SWE-Bench Verified | reasoning frontend_development code | 72.5 | 来源 |
| TAU-bench Airline | reasoning communication | 59.6 | 来源 |
| Terminal-bench | reasoning code | 39.2 | 来源 |
| ARC-AGI v2 | reasoning vision spatial_reasoning | 8.6 | 来源 |
Pricing
API 价格对比
| 服务商 | 输入价 | 输出价 | 上下文 | 吞吐(tok/s) | 延迟(s) | 函数调用 | 代码执行 | 联网搜索 |
|---|---|---|---|---|---|---|---|---|
| Jiekou.AI | $13.50 | $67.50 | 200K | — | — | ✓ | ✗ | ✗ |
| NanoGPT | $14.99 | $75.00 | 200K | — | — | ✓ | ✗ | ✗ |
| 302.AI | $15.00 | $75.00 | 200K | — | — | ✓ | ✗ | ✗ |
| Abacus | $15.00 | $75.00 | 200K | — | — | ✓ | ✗ | ✗ |
价格单位：美元/百万 token，数据来自社区整理，仅供参考。