评测来源/Arena Agent Overall
观察中专业工作
Arena Agent Overall
LMArena · Arena Agent Overall 的公开模型评测结果。
本来源的模型成绩
| Claude Opus 5.5 | 0.20144840119751856 | 90.3 | 2026-11-04 |
| Claude Sonnet 5.5 | 0.15423674759515776 | 80.8 | 2026-11-04 |
| GPT-6 Astra | 0.1469702596284481 | 79.4 | 2026-11-04 |
| Claude Fable 5.1 | 0.13857005028518268 | 77.7 | 2026-11-04 |
| GPT-6.1 Sol | 0.12513945767905282 | 75.0 | 2026-11-04 |
| GPT-6.1 Terra | 0.11396048113208619 | 72.8 | 2026-11-04 |
| Gemini 4 Argon | 0.10525164824272239 | 71.1 | 2026-11-04 |
| Muse Wonder Full | 0.09422636744409608 | 68.8 | 2026-11-04 |
| Claude Fable 5 | 0.078224963210975 | 65.6 | 2026-11-04 |
| Claude Opus 5 | 0.07661953023610667 | 65.3 | 2026-11-04 |
| Claude Opus 4.8 | 0.07131379676560229 | 64.3 | 2026-10-08 |
| GPT-6 Sol | 0.06778131892924635 | 63.6 | 2026-11-04 |
| Muse Wonder | 0.049356218746185024 | 59.9 | 2026-10-16 |
| Muse Wonder | 0.049356218746185024 | 59.9 | 2026-10-16 |
| GPT-5.6 Sol | 0.04572267729564615 | 59.1 | 2026-11-04 |
| GPT-5.5 | 0.03120313955899052 | 56.2 | 2026-10-20 |
| MiMo V2.6 Pro | 0.026810364256876926 | 55.4 | 2026-11-04 |
| Grok 4.7 | 0.02680192300589593 | 55.4 | 2026-11-04 |
| MiMo V2.6 Flash | 0.02598333419739363 | 55.2 | 2026-10-16 |
| MiMo V2.6 Flash | 0.02598333419739363 | 55.2 | 2026-10-16 |
| Gemini 3.8 Flash | 0.025819514653005995 | 55.2 | 2026-11-04 |
| Muse Spark 1.3 | 0.025113832728772198 | 55.0 | 2026-11-04 |
| Claude Sonnet 5 | 0.024429680071076675 | 54.9 | 2026-11-04 |
| Muse Wonder | 0.024335809889644497 | 54.9 | 2026-11-04 |
| DeepSeek V4.1 Flash | 0.019121718092652323 | 53.8 | 2026-10-28 |
| GPT-6 Luna | 0.018937239039092082 | 53.8 | 2026-11-04 |
| Claude Haiku 5.5 | 0.014944374231666368 | 53.0 | 2026-10-16 |
| Claude Haiku 5.5 | 0.014944374231666368 | 53.0 | 2026-10-16 |
| GPT-5.4 | 0.01359757877358846 | 52.7 | 2026-10-20 |
| Grok 4.5 | 0.011942064565826428 | 52.4 | 2026-10-08 |
| GLM 5.3 | 0.010890920338008515 | 52.2 | 2026-11-04 |
| Kimi K3 | 0.008419453885193499 | 51.7 | 2026-11-04 |
| GLM-5.2 | 0.0063863950447943145 | 51.3 | 2026-11-04 |
| Claude Haiku 5.5 | 0.006244188898989065 | 51.2 | 2026-11-04 |
| Qwen3.8 Max | 0.004788293142734283 | 51.0 | 2026-11-04 |
| Grok 4.6 | 0.004542626115501247 | 50.9 | 2026-11-04 |
| MiMo V2.6 Flash | 0.002000589395452135 | 50.4 | 2026-11-04 |
| GLM-5.3 Flash | -0.011620853988777428 | 47.7 | 2026-10-16 |
| GLM-5.3 Flash | -0.011620853988777428 | 47.7 | 2026-10-16 |
| GPT-5.6 Terra | -0.01504690834577408 | 47.0 | 2026-11-04 |
| GPT-5.6 Luna | -0.023322547362270542 | 45.3 | 2026-10-20 |
| GLM-5.3 Flash | -0.023939650496158076 | 45.2 | 2026-11-04 |
| MiniMax M3.1 Flash | -0.02843512991092605 | 44.3 | 2026-11-04 |
| Step 5 Preview | -0.028761238788302872 | 44.2 | 2026-11-04 |
| Gemini 3.1 Pro Preview | -0.044409818497287236 | 41.1 | 2026-11-04 |
| Mistral Large 4 | -0.05462655237209272 | 39.1 | 2026-11-04 |
| Qwen3.7 Max | -0.0651996046174915 | 37.0 | 2026-11-04 |
| Qwen3.7 Plus | -0.07459946086080223 | 35.1 | 2026-11-04 |
| Muse Spark 1.1 | -0.07659459536438275 | 34.7 | 2026-11-04 |
| Muse Spark 1.2 | -0.07698988459758775 | 34.6 | 2026-11-04 |
| Hy3 | -0.0771969842666558 | 34.6 | 2026-11-04 |
| Gemini 3.6 Flash | -0.0779187096208923 | 34.4 | 2026-10-08 |
| MiMo-V2.5-Pro | -0.09027032879654748 | 31.9 | 2026-11-04 |
| MiniMax-M3 | -0.10396707055734586 | 29.2 | 2026-11-04 |
| NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4 | -0.1593210465176367 | 18.1 | 2026-11-04 |
| Inkling | -0.16332138597542212 | 17.3 | 2026-11-04 |
| Inkling Small | -0.1703963452049372 | 15.9 | 2026-11-04 |