- 作者
- DoDoT Data Lab
- Provider
- LMArena
- 交付方式
- 站内运行
LMArena 社区评比
大模型排行榜
根据匿名双模型对战生成的 Arena Score。分数反映社区偏好，不代表事实准确率、安全性或绝对能力。
| 名次 | 模型 | 机构 | Arena Score | 置信区间 | 票数 | 许可证 |
|---|---|---|---|---|---|---|
| #1 | Claude Fable 5claude-fable-5 | anthropic | 1,507 | 1,502–1,513 | 19,390 | Proprietary |
| #2 | Claude Opus 4.6 Thinkingclaude-opus-4-6-thinking | anthropic | 1,505 | 1,501–1,508 | 69,336 | Proprietary |
| #3 | Claude Opus 4.7 Thinkingclaude-opus-4-7-thinking | anthropic | 1,502 | 1,498–1,506 | 57,018 | Proprietary |
| #4 | Muse Spark 1.2 (xHigh)muse-spark-1.2 (xHigh) | meta | 1,498 | 1,485–1,511 | 2,057 | Proprietary |
| #5 | Claude Opus 4.6claude-opus-4-6 | anthropic | 1,497 | 1,494–1,501 | 73,158 | Proprietary |
| #6 | Qwen 3.8 Maxqwen3.8-max | alibaba | 1,497 | 1,488–1,506 | 4,662 | Proprietary |
| #7 | Claude Opus 4.7claude-opus-4-7 | anthropic | 1,493 | 1,489–1,497 | 58,135 | Proprietary |
| #8 | Claude Opus 5 Highclaude-opus-5-high | anthropic | 1,493 | 1,487–1,499 | 14,902 | Proprietary |
| #9 | Claude Opus 5 Maxclaude-opus-5-max | anthropic | 1,488 | 1,481–1,496 | 7,137 | Proprietary |
| #10 | Muse Sparkmuse-spark | meta | 1,488 | 1,482–1,494 | 13,476 | Proprietary |
| #11 | Muse Spark 1.1muse-spark-1.1 | meta | 1,487 | 1,482–1,493 | 14,227 | Proprietary |
| #12 | Gemini 3.1 Pro Previewgemini-3.1-pro-preview | 1,487 | 1,483–1,490 | 91,328 | Proprietary | |
| #13 | Gemini 3 Progemini-3-pro | 1,486 | 1,482–1,489 | 41,242 | Proprietary | |
| #14 | Kimi K3 Maxkimi-k3-max | moonshot | 1,485 | 1,479–1,492 | 9,861 | Kimi K3 license |
| #15 | Gemini 3.6 Flashgemini-3.6-flash | 1,485 | 1,479–1,491 | 10,968 | Proprietary | |
| #16 | Claude Opus 4.8 Thinkingclaude-opus-4-8-thinking | anthropic | 1,483 | 1,478–1,488 | 37,546 | Proprietary |
| #17 | GPT-5.5 Highgpt-5.5-high | openai | 1,482 | 1,478–1,486 | 52,142 | Proprietary |
| #18 | GPT-5.6 Sol xHighgpt-5.6-sol-xhigh | openai | 1,482 | 1,476–1,488 | 12,839 | Proprietary |
| #19 | GPT-5.4 Highgpt-5.4-high | openai | 1,477 | 1,473–1,481 | 60,156 | Proprietary |
| #20 | GPT-5.5gpt-5.5 | openai | 1,477 | 1,473–1,481 | 53,407 | Proprietary |
| #21 | Gemini 3.5 Flash Highgemini-3.5-flash-high | 1,476 | 1,471–1,481 | 22,797 | Proprietary | |
| #22 | GPT-5.2 Chat Latest 20260210gpt-5.2-chat-latest-20260210 | openai | 1,476 | 1,472–1,480 | 34,020 | Proprietary |
| #23 | Qwen 3.7 Max Previewqwen3.7-max-preview | alibaba | 1,475 | 1,465–1,485 | 3,695 | Proprietary |
| #24 | Grok 4.20 Beta1grok-4.20-beta1 | xai | 1,474 | 1,470–1,479 | 26,553 | Proprietary |
| #25 | Gemini 3.5 Flash Mediumgemini-3.5-flash-medium | 1,474 | 1,469–1,480 | 21,100 | Proprietary | |
| #26 | GPT-5.5 Instantgpt-5.5-instant | openai | 1,474 | 1,468–1,479 | 25,642 | Proprietary |
| #27 | Gemini 3 Flashgemini-3-flash | 1,473 | 1,468–1,477 | 30,625 | Proprietary | |
| #28 | Claude Opus 4.8claude-opus-4-8 | anthropic | 1,473 | 1,468–1,477 | 38,131 | Proprietary |
| #29 | Claude Opus 4.5 20251101 Thinking 32Kclaude-opus-4-5-20251101-thinking-32k | anthropic | 1,473 | 1,469–1,477 | 36,960 | Proprietary |
| #30 | Claude Sonnet 4.6claude-sonnet-4-6 | anthropic | 1,472 | 1,468–1,476 | 63,453 | Proprietary |
| #31 | Grok 4.20 Beta 0309 Reasoninggrok-4.20-beta-0309-reasoning | xai | 1,472 | 1,468–1,476 | 61,639 | Proprietary |
| #32 | Grok 4.20 Multi Agent Beta 0309grok-4.20-multi-agent-beta-0309 | xai | 1,471 | 1,467–1,475 | 60,322 | Proprietary |
| #33 | GLM-5.2 Maxglm-5.2-max | zai | 1,471 | 1,466–1,476 | 24,250 | MIT |
| #34 | Claude Opus 4.5 20251101claude-opus-4-5-20251101 | anthropic | 1,469 | 1,466–1,472 | 70,504 | Proprietary |
| #35 | Grok 4.5grok-4.5 | xai | 1,468 | 1,463–1,474 | 14,883 | Proprietary |
| #36 | GLM-5.1glm-5.1 | zai | 1,468 | 1,464–1,472 | 36,716 | MIT |
| #37 | ERNIE 5.1ernie-5.1 | baidu | 1,467 | 1,463–1,472 | 36,820 | Proprietary |
| #38 | GPT-5.6 Terra xHighgpt-5.6-terra-xhigh | openai | 1,467 | 1,461–1,473 | 13,275 | Proprietary |
| #39 | MiMo V2.5 Promimo-v2.5-pro | xiaomi | 1,467 | 1,463–1,471 | 47,805 | MIT |
| #40 | Grok 4.1 Thinkinggrok-4.1-thinking | xai | 1,466 | 1,463–1,469 | 65,044 | Proprietary |
| #41 | Qwen 3.5 Max Previewqwen3.5-max-preview | alibaba | 1,465 | 1,460–1,470 | 21,274 | Proprietary |
| #42 | GPT-5.4gpt-5.4 | openai | 1,465 | 1,461–1,469 | 63,097 | Proprietary |
| #43 | Claude Sonnet 5 Highclaude-sonnet-5-high | anthropic | 1,463 | 1,457–1,468 | 20,311 | Proprietary |
| #44 | Kimi K2.6kimi-k2.6 | moonshot | 1,461 | 1,456–1,465 | 37,236 | Modified MIT |
| #45 | Qwen 3.6 Max Previewqwen3.6-max-preview | alibaba | 1,460 | 1,452–1,469 | 5,146 | Proprietary |
| #46 | Grok 4.1grok-4.1 | xai | 1,460 | 1,456–1,463 | 67,246 | Proprietary |
| #47 | Gemini 3.5 Flash Litegemini-3.5-flash-lite | 1,459 | 1,452–1,465 | 10,847 | Proprietary | |
| #48 | Gemini 3 Flash (thinking Minimal)gemini-3-flash (thinking-minimal) | 1,459 | 1,455–1,462 | 85,546 | Proprietary | |
| #49 | Qwen 3.7 Plusqwen3.7-plus | alibaba | 1,458 | 1,453–1,463 | 28,111 | Proprietary |
| #50 | DeepSeek V4 Prodeepseek-v4-pro | deepseek | 1,457 | 1,453–1,461 | 51,266 | MIT |
| #51 | GLM-5glm-5 | zai | 1,457 | 1,452–1,461 | 27,591 | MIT |
| #52 | dola-seed-2.0-pro | bytedance | 1,456 | 1,452–1,459 | 73,874 | Proprietary |
| #53 | Claude Sonnet 4.5 20250929 Thinking 32Kclaude-sonnet-4-5-20250929-thinking-32k | anthropic | 1,456 | 1,453–1,458 | 81,888 | Proprietary |
| #54 | DeepSeek V4 Pro High Previewdeepseek-v4-pro-high-preview | deepseek | 1,456 | 1,451–1,460 | 48,756 | MIT |
| #55 | Claude Sonnet 4.5 20250929claude-sonnet-4-5-20250929 | anthropic | 1,455 | 1,452–1,458 | 80,284 | Proprietary |
| #56 | GPT-5.1 Highgpt-5.1-high | openai | 1,454 | 1,451–1,458 | 40,680 | Proprietary |
| #57 | hy3 | tencent | 1,453 | 1,444–1,463 | 3,941 | Apache 2.0 |
| #58 | Gemma 4 31bgemma-4-31b | 1,451 | 1,443–1,459 | 5,839 | Apache 2.0 | |
| #59 | Kimi K2.5 Thinkingkimi-k2.5-thinking | moonshot | 1,451 | 1,447–1,454 | 68,179 | Modified MIT |
| #60 | GPT-5.6 Luna xHighgpt-5.6-luna-xhigh | openai | 1,450 | 1,444–1,456 | 13,651 | Proprietary |
| #61 | Claude Opus 4.1 20250805 Thinking 16Kclaude-opus-4-1-20250805-thinking-16k | anthropic | 1,449 | 1,446–1,453 | 49,720 | Proprietary |
| #62 | ERNIE 5.0 Preview 1203ernie-5.0-preview-1203 | baidu | 1,449 | 1,442–1,455 | 9,727 | Proprietary |
| #63 | GPT-5.3 Chat Latestgpt-5.3-chat-latest | openai | 1,449 | 1,444–1,453 | 32,583 | Proprietary |
| #64 | GPT-5.4 Mini Highgpt-5.4-mini-high | openai | 1,448 | 1,444–1,452 | 58,984 | Proprietary |
| #65 | MiMo V2 Promimo-v2-pro | xiaomi | 1,448 | 1,443–1,453 | 24,211 | Proprietary |
| #66 | Claude Opus 4.1 20250805claude-opus-4-1-20250805 | anthropic | 1,447 | 1,444–1,450 | 77,192 | Proprietary |
| #67 | ERNIE 5.0 0110ernie-5.0-0110 | baidu | 1,447 | 1,443–1,451 | 35,015 | Proprietary |
| #68 | Gemini 2.5 Progemini-2.5-pro | 1,446 | 1,443–1,448 | 124,014 | Proprietary | |
| #69 | GPT-4.5 Preview 2025 02 27gpt-4.5-preview-2025-02-27 | openai | 1,445 | 1,439–1,450 | 14,547 | Proprietary |
| #70 | MiniMax M3minimax-m3 | minimax | 1,445 | 1,440–1,449 | 34,343 | MiniMax Community License |
| #71 | Qwen 3.6 Plusqwen3.6-plus | alibaba | 1,443 | 1,439–1,448 | 45,033 | Proprietary |
| #72 | ChatGPT-4o Latest 20250326chatgpt-4o-latest-20250326 | openai | 1,443 | 1,440–1,446 | 82,353 | Proprietary |
| #73 | Qwen 3.5 397b A17bqwen3.5-397b-a17b | alibaba | 1,442 | 1,439–1,446 | 63,719 | Apache 2.0 |
| #74 | GLM-4.7glm-4.7 | zai | 1,442 | 1,436–1,448 | 12,077 | MIT |
| #75 | Grok 4.3grok-4.3 | xai | 1,442 | 1,438–1,446 | 53,234 | Proprietary |
| #76 | inkling | thinky | 1,442 | 1,435–1,448 | 12,086 | Apache 2.0 |
| #77 | GPT-5.1gpt-5.1 | openai | 1,439 | 1,435–1,442 | 43,306 | Proprietary |
| #78 | Gemma 4 26b A4bgemma-4-26b-a4b | 1,438 | 1,431–1,446 | 5,755 | Apache 2.0 | |
| #79 | DeepSeek V4 Flash High Previewdeepseek-v4-flash-high-preview | deepseek | 1,438 | 1,433–1,442 | 48,316 | MIT |
| #80 | GPT-5.2 Highgpt-5.2-high | openai | 1,438 | 1,434–1,441 | 47,621 | Proprietary |
| #81 | DeepSeek V4 Flashdeepseek-v4-flash | deepseek | 1,436 | 1,432–1,440 | 48,561 | MIT |
| #82 | GPT-5.2gpt-5.2 | openai | 1,435 | 1,432–1,438 | 78,662 | Proprietary |
| #83 | longcat-flash-chat-2602-exp | meituan | 1,435 | 1,430–1,440 | 27,706 | Proprietary |
| #84 | Qwen 3 Max Previewqwen3-max-preview | alibaba | 1,435 | 1,430–1,439 | 27,670 | Proprietary |
| #85 | MiMo V2.5mimo-v2.5 | xiaomi | 1,434 | 1,430–1,438 | 44,200 | MIT |
| #86 | GPT-5 Highgpt-5-high | openai | 1,434 | 1,429–1,438 | 31,877 | Proprietary |
| #87 | glm-5v-turbo | zai | 1,433 | 1,426–1,440 | 9,288 | Proprietary |
| #88 | Gemini 3.1 Flash Lite Previewgemini-3.1-flash-lite-preview | 1,432 | 1,428–1,436 | 60,041 | Proprietary | |
| #89 | Kimi K2.5 Instantkimi-k2.5-instant | moonshot | 1,431 | 1,425–1,438 | 8,138 | Modified MIT |
| #90 | Grok 4 1 Fast Reasoninggrok-4-1-fast-reasoning | xai | 1,431 | 1,428–1,434 | 56,402 | Proprietary |
| #91 | o3-2025-04-16 | openai | 1,431 | 1,427–1,434 | 59,661 | Proprietary |
| #92 | MiMo V2 Omnimimo-v2-omni | xiaomi | 1,430 | 1,424–1,435 | 19,216 | Proprietary |
| #93 | Kimi K2 Thinking Turbokimi-k2-thinking-turbo | moonshot | 1,430 | 1,427–1,433 | 61,677 | Modified MIT |
| #94 | mistral-medium-3.5 | mistral | 1,427 | 1,421–1,434 | 10,887 | Modified MIT |
| #95 | GPT-5 Chatgpt-5-chat | openai | 1,427 | 1,423–1,431 | 31,517 | Proprietary |
| #96 | nvidia-nemotron-3-ultra-550b-a55b-nvfp4 | nvidia | 1,426 | 1,419–1,433 | 10,564 | OpenMDW-1.1 |
| #97 | amazon-nova-experimental-chat-26-02-10 | amazon | 1,426 | 1,416–1,436 | 3,392 | Proprietary |
| #98 | DeepSeek V3.2deepseek-v3.2 | deepseek | 1,425 | 1,421–1,429 | 46,999 | MIT |
| #99 | GLM-4.6glm-4.6 | zai | 1,425 | 1,421–1,429 | 35,589 | MIT |
| #100 | DeepSeek V3.2 Exp Thinkingdeepseek-v3.2-exp-thinking | deepseek | 1,425 | 1,418–1,431 | 9,051 | MIT |
