大幅降价、无限聊天、编码能力超越人类专家,Claude Opus 4.5重夺最强模型王冠
3 6 Ke·2025-11-25 01:48

| | Opus 4.5 | Sonnet 4.5 | Opus 4.1 | Gemini 3 Pro | GPT-5.1 | | --- | --- | --- | --- | --- | --- | | Agentic coding | | | | | 76.3% | | SWE-bench Verlfied | 80.9% | 77.2% | 74.5% | 76.2% | 77.9% | | | | | | | Cadea-Max | | Agentic terminal | | | | | 47.6% | | coding | 59.3% | 50.0% | 46.5% | 54.2% | 58.1% | | Terminal-bench 2.0 | | | | | Cochia Max | | | Recal | frisk | lickel | Retal | | | | 88.9% | 86.2% | 86.8% | 85.3% | - | | Agentic tool use | | | | | | | t2-bench | Telecom | Telecore | Telecom | Te ...