Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents
发布日期2026年5月28日
参数规模—
上下文长度1.0M
许可证Proprietary
知识截止2026年1月1日
Benchmarks
评测成绩
| 评测基准 | 类别 | 分数 | 来源 |
|---|---|---|---|
| SWE-Bench Verified | reasoning frontend_development code | 88.6 | 来源 |
| OSWorld-Verified | general | 83.4 | 来源 |
| Terminal-Bench | reasoning code | 74.6 | 来源 |
| SWE-Bench Pro | code | 69.2 | 来源 |
| Humanity's Last Exam | general | 57.9 | 来源 |
| Humanity's Last Exam | general | 49.8 | 来源 |
| FrontierCode | code | 13.4 | 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 |
|---|---|---|---|---|---|---|---|---|
| Kenari | $0.00 | $0.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| UnoRouter | $0.43 | $2.13 | 1.0M | — | — | ✓ | ✗ | ✗ |
| Abacus | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| AIHubMix | $5.00 | $25.00 | 200K | — | — | ✓ | ✗ | ✗ |
| Anthropic | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| Azure | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| Azure Cognitive Services | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| DaoXE | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| FreeModel | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| LLM Gateway | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| Neon | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| OpenCode Zen | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| Pioneer | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| routing.run | $5.00 | $25.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| Venice AI | $6.00 | $30.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
价格单位：美元/百万 token，数据来自社区整理，仅供参考。