高级工具调用能力
搜索文档
Claude Opus 4.5夺回编程王座,超Gemini 3 Pro和GPT-5.1
AI前线· 2025-11-25 13:03
目前测试版(Beta 版)已上线,开发者可直接通过 Claude API 调用。 | | Opus 4.5 | Sonnet 4.5 | Opus 4.1 | Gemini 3 Pro | GPT-5.1 | | | --- | --- | --- | --- | --- | --- | --- | | Agentic coding | | | | | 76.3% | | | SWE-bench Verified | 80.9% | 77.2% | 74.5% | 76.2% | 77.9% | | | | | | | | Codex-Max | | | Agentic terminal | | | | | 47.6% | | | coding | 59.3% | 50.0% | 46.5% | 54.2% | | | | Terminal-bench 2.0 | | | | | 58.1% | | | | | | | | Codex-Max | | | | Retail | Retail | Retail | Retail | - | | | Agentic tool use | 88.9% | 86.2% ...