模型排行
| 排名 | 模型 | 机构 | 分数 | 置信区间 | 开放状态 | 证据 |
|---|---|---|---|---|---|---|
| 1 | claude-fable-5 Arena rating ↑ | anthropic | 1,326 | 0–0 | Proprietary | 11,304 票 来源 |
| 2 | claude-fable-5.1-max Arena rating ↑ | anthropic | 1,322 | 0–0 | Proprietary | 2,701 票 来源 |
| 3 | claude-opus-5-high Arena rating ↑ | anthropic | 1,321 | 0–0 | Proprietary | 11,599 票 来源 |
| 4 | claude-opus-4-7 Arena rating ↑ | anthropic | 1,318 | 0–0 | Proprietary | 21,450 票 来源 |
| 5 | claude-opus-4-7-high Arena rating ↑ | anthropic | 1,316 | 0–0 | Proprietary | 21,092 票 来源 |
| 6 | qwen3.8-max Arena rating ↑ | alibaba | 1,315 | 0–0 | Proprietary | 8,665 票 来源 |
| 7 | claude-opus-4-6-high Arena rating ↑ | anthropic | 1,315 | 0–0 | Proprietary | 20,835 票 来源 |
| 8 | muse-spark-1.3-max Arena rating ↑ | meta | 1,315 | 0–0 | Proprietary | 1,804 票 来源 |
| 9 | claude-opus-4-6 Arena rating ↑ | anthropic | 1,312 | 0–0 | Proprietary | 25,140 票 来源 |
| 10 | gemini-3.5-flash-high Arena rating ↑ | 1,310 | 0–0 | Proprietary | 9,579 票 来源 | |
| 11 | gemini-3.5-flash-medium Arena rating ↑ | 1,307 | 0–0 | Proprietary | 9,618 票 来源 | |
| 12 | muse-spark Arena rating ↑ | meta | 1,306 | 0–0 | Proprietary | 5,572 票 来源 |
| 13 | gemini-3-pro Arena rating ↑ | 1,304 | 0–0 | Proprietary | 12,995 票 来源 | |
| 14 | muse-spark-1.2 (xHigh) Arena rating ↑ | meta | 1,304 | 0–0 | Proprietary | 1,841 票 来源 |
| 15 | gpt-5.4-high Arena rating ↑ | openai | 1,302 | 0–0 | Proprietary | 25,357 票 来源 |
| 16 | gemini-3.6-flash-high Arena rating ↑ | 1,300 | 0–0 | Proprietary | 5,159 票 来源 | |
| 17 | glm-5.3-flash Arena rating ↑ | zai | 1,299 | 0–0 | MIT | 3,110 票 来源 |
| 18 | gpt-5.5 Arena rating ↑ | openai | 1,298 | 0–0 | Proprietary | 23,423 票 来源 |
| 19 | gemini-3.1-pro-preview Arena rating ↑ | 1,296 | 0–0 | Proprietary | 40,691 票 来源 | |
| 20 | gpt-5.5-high Arena rating ↑ | openai | 1,294 | 0–0 | Proprietary | 21,960 票 来源 |
| 21 | muse-spark-1.1 Arena rating ↑ | meta | 1,294 | 0–0 | Proprietary | 8,447 票 来源 |
| 22 | gpt-5.4 Arena rating ↑ | openai | 1,293 | 0–0 | Proprietary | 21,188 票 来源 |
| 23 | claude-opus-4-8-high Arena rating ↑ | anthropic | 1,292 | 0–0 | Proprietary | 15,274 票 来源 |
| 24 | grok-4.5 Arena rating ↑ | xai | 1,289 | 0–0 | Proprietary | 8,355 票 来源 |
| 25 | claude-opus-4-8 Arena rating ↑ | anthropic | 1,288 | 0–0 | Proprietary | 15,801 票 来源 |
| 26 | gemini-3-flash Arena rating ↑ | 1,285 | 0–0 | Proprietary | 36,679 票 来源 | |
| 27 | claude-sonnet-4-6 Arena rating ↑ | anthropic | 1,283 | 0–0 | Proprietary | 25,552 票 来源 |
| 28 | kimi-k2.6 Arena rating ↑ | moonshot | 1,280 | 0–0 | Modified MIT | 15,347 票 来源 |
| 29 | gpt-5.6-sol-xhigh Arena rating ↑ | openai | 1,280 | 0–0 | Proprietary | 7,729 票 来源 |
| 30 | gpt-6-astra-max Arena rating ↑ | openai | 1,279 | 0–0 | Proprietary | 1,367 票 来源 |
| 31 | qwen3.7-plus Arena rating ↑ | alibaba | 1,278 | 0–0 | Proprietary | 10,656 票 来源 |
| 32 | claude-sonnet-5-high Arena rating ↑ | anthropic | 1,277 | 0–0 | Proprietary | 10,289 票 来源 |
| 33 | gemma-4-31b Arena rating ↑ | 1,276 | 0–0 | Apache 2.0 | 35,713 票 来源 | |
| 34 | dola-seed-2.0-pro Arena rating ↑ | bytedance | 1,275 | 0–0 | Proprietary | 10,280 票 来源 |
| 35 | qwen3.8-27b Arena rating ↑ | alibaba | 1,272 | 0–0 | Apache 2.0 | 4,577 票 来源 |
| 36 | gpt-5.6-terra-xhigh Arena rating ↑ | openai | 1,270 | 0–0 | Proprietary | 7,726 票 来源 |
| 37 | gemini-3.5-flash-lite Arena rating ↑ | 1,268 | 0–0 | Proprietary | 5,254 票 来源 | |
| 38 | gpt-5.2-chat-latest-20260210 Arena rating ↑ | openai | 1,268 | 0–0 | Proprietary | 15,432 票 来源 |
| 39 | kimi-k2.5-thinking Arena rating ↑ | moonshot | 1,267 | 0–0 | Modified MIT | 26,646 票 来源 |
| 40 | grok-4.6-high Arena rating ↑ | xai | 1,267 | 0–0 | Proprietary | 3,632 票 来源 |
| 41 | gemini-3-flash (thinking-minimal) Arena rating ↑ | 1,265 | 0–0 | Proprietary | 34,860 票 来源 | |
| 42 | qwen3.5-397b-a17b Arena rating ↑ | alibaba | 1,265 | 0–0 | Apache 2.0 | 27,569 票 来源 |
| 43 | grok-4.20-beta-0309-reasoning Arena rating ↑ | xai | 1,263 | 0–0 | Proprietary | 25,123 票 来源 |
| 44 | glm-5v-turbo Arena rating ↑ | zai | 1,262 | 0–0 | Proprietary | 37,424 票 来源 |
| 45 | gemini-2.5-pro Arena rating ↑ | 1,262 | 0–0 | Proprietary | 87,635 票 来源 | |
| 46 | grok-4.20-multi-agent-beta-0309 Arena rating ↑ | xai | 1,260 | 0–0 | Proprietary | 22,925 票 来源 |
| 47 | gemma-4-26b-a4b Arena rating ↑ | 1,259 | 0–0 | Apache 2.0 | 22,658 票 来源 | |
| 48 | gpt-5.6-luna-xhigh Arena rating ↑ | openai | 1,259 | 0–0 | Proprietary | 7,793 票 来源 |
| 49 | kimi-k2.5-instant Arena rating ↑ | moonshot | 1,255 | 0–0 | Modified MIT | 3,872 票 来源 |
| 50 | minimax-m3 Arena rating ↑ | minimax | 1,254 | 0–0 | MiniMax Community License | 15,192 票 来源 |