LMArena Coding Arena 代码能力排行榜
基于 LMArena Coding Arena 用户匿名投票的最新AI大模型代码编程能力排行榜，涵盖各模型的 Elo 得分、95% 置信区间、投票量、机构与许可证。
榜首模型
gemini-4-argon-high
最高得分
1559.00
模型数量
405
数据版本
2026年09月30日
数据来源: LM Arena
关于本排行榜
本排行榜展示了当前 AI 大模型在代码编程任务中的实力排名。数据来源于 LMArena （前身为 LMSYS Chatbot Arena）的 Coding 子赛道，通过真实用户匿名盲测投票评估各模型在代码编程任务中的表现。
评测方法概要
匿名盲测：用户发出编程问题后，由两个"隐藏身份"的模型分别给出代码解答，用户投票选出更好的回答，排除品牌偏见。
Elo 评分：采用 Bradley-Terry 模型计算 Elo 分数，分数越高说明该模型的代码回答越容易被用户选择。
覆盖多种编程场景：包括代码生成、Bug 修复、算法实现、代码解释等高频真实编程场景。
DataLearner 在原始数据基础上提供中文解读与深度分析，并将排行榜模型关联至 DataLearner 模型库，方便您一键查看模型详情、API 定价、评测得分等完整信息。
排名总表
| 排名 | 模型名称 | 得分 | 95% CI | 投票数 | 机构 | 许可证 |
|---|---|---|---|---|---|---|
 gemini-4-argon-highGoogle | 1559.00 | +/-17 | 1,225 | Proprietary | ||
 claude-fable-5-highAnthropic | 1552.00 | +/-7 | 9,946 | Anthropic | Proprietary | |
| Claude Opus 4.6 (high)Anthropic | 1551.00 | +/-6 | 20,144 | Anthropic | Proprietary | |
| 4 | Opus 4.7 (high)Anthropic | 1551.00 | +/-6 | 18,437 | Anthropic | Proprietary |
| 5 | Claude Opus 4.6Anthropic | 1547.00 | +/-5 | 22,791 | Anthropic | Proprietary |
| 6 | Opus 4.7Anthropic | 1547.00 | +/-6 | 18,642 | Anthropic | Proprietary |
| 7 |  kimi-k3-maxMoonshot | 1541.00 | +/-8 | 7,165 | Moonshot | Kimi K3 license |
| 8 | mimo-v2.6-proXiaomi | 1540.00 | +/-18 | 1,101 | Xiaomi | MIT |
| 9 |  gpt-6-astra-maxOpenAI | 1539.00 | +/-13 | 2,010 | OpenAI | Proprietary |
| 10 |  muse-spark-1.3-maxMeta | 1539.00 | +/-11 | 3,142 | Meta | Proprietary |
| 11 | claude-opus-5.5-highAnthropic | 1538.00 | +/-19 | 918 | Anthropic | Proprietary |
| 12 | Muse Spark 1.2 (xhigh)Facebook AI研究实验室 | 1535.00 | +/-18 | 1,085 | Facebook AI研究实验室 | Proprietary |
| 13 | Muse Spark 1.1Facebook AI研究实验室 | 1534.00 | +/-7 | 10,381 | Facebook AI研究实验室 | Proprietary |
| 14 | GPT-5.6 Sol (xhigh)OpenAI | 1534.00 | +/-7 | 9,788 | OpenAI | Proprietary |
| 15 | claude-opus-5-highAnthropic | 1534.00 | +/-6 | 15,273 | Anthropic | Proprietary |
| 16 | Claude Opus 4.8 (high)Anthropic | 1533.00 | +/-6 | 17,275 | Anthropic | Proprietary |
| 17 | Claude Fable 5.1 (max)Anthropic | 1532.00 | +/-13 | 2,254 | Anthropic | Proprietary |
| 18 | gemini-3.8-flash-highGoogle | 1531.00 | +/-8 | 6,660 | Proprietary | |
| 19 | Opus 4.5Anthropic | 1531.00 | +/-7 | 7,790 | Anthropic | Proprietary |
| 20 | Claude Opus 4.8Anthropic | 1530.00 | +/-6 | 17,933 | Anthropic | Proprietary |
| 21 | claude-opus-5-maxAnthropic | 1530.00 | +/-8 | 7,234 | Anthropic | Proprietary |
| 22 | Muse SparkFacebook AI研究实验室 | 1530.00 | +/-10 | 3,930 | Facebook AI研究实验室 | Proprietary |
| 23 |  deepseek-v4.1-flash-maxDeepSeek | 1529.00 | +/-12 | 2,314 | DeepSeek | MIT |
| 24 | Claude Sonnet 4.6Anthropic | 1529.00 | +/-6 | 19,706 | Anthropic | Proprietary |
| 25 |  Qwen3.7-Max-Preview阿里巴巴 | 1524.00 | +/-18 | 1,167 | 阿里巴巴 | Proprietary |
| 26 |  glm-5.3-flashZ.ai | 1523.00 | +/-9 | 5,833 | Z.ai | MIT |
| 27 | Claude Opus 4Anthropic | 1523.00 | +/-5 | 17,972 | Anthropic | Proprietary |
| 28 | gpt-6-sol-maxOpenAI | 1523.00 | +/-14 | 1,799 | OpenAI | Proprietary |
| 29 | qwen3.8-maxAlibaba | 1522.00 | +/-8 | 6,480 | Alibaba | Proprietary |
| 30 | mimo-v2.6-flashXiaomi | 1521.00 | +/-16 | 1,361 | Xiaomi | MIT |
| 31 | mimo-v2.5-proXiaomi | 1521.00 | +/-6 | 19,204 | Xiaomi | MIT |
| 32 | glm-5.3-maxZ.ai | 1521.00 | +/-10 | 4,283 | Z.ai | MIT |
| 33 | Gemini 3.1 Pro PreviewGoogle DeepMind | 1521.00 | +/-5 | 33,447 | Google DeepMind | Proprietary |
| 34 | GPT-5.4 (high)OpenAI | 1520.00 | +/-6 | 17,560 | OpenAI | Proprietary |
| 35 | Claude Sonnet 4.5Anthropic | 1519.00 | +/-5 | 19,909 | Anthropic | Proprietary |
| 36 | Claude Sonnet 4.5 (high)Anthropic | 1519.00 | +/-6 | 12,249 | Anthropic | Proprietary |
| 37 | GPT-5.5 (high)OpenAI | 1519.00 | +/-6 | 19,151 | OpenAI | Proprietary |
| 38 | gemini-3.6-flash-highGoogle | 1518.00 | +/-7 | 10,226 | Proprietary | |
| 39 | gemini-3.7-flash-highGoogle | 1518.00 | +/-8 | 5,873 | Proprietary | |
| 40 | Gemini 3 ProGoogle DeepMind | 1518.00 | +/-7 | 8,788 | Google DeepMind | Proprietary |
| 41 | Kimi K2.6Moonshot AI | 1516.00 | +/-7 | 11,069 | Moonshot AI | Modified MIT |
| 42 | GPT-5.6 Terra (xhigh)OpenAI | 1515.00 | +/-7 | 10,012 | OpenAI | Proprietary |
| 43 | GPT-5.2 Chat (0210)OpenAI | 1515.00 | +/-7 | 9,630 | OpenAI | Proprietary |
| 44 |  DOLA Seed 2.0 Pro字节跳动Seed团队 | 1514.00 | +/-5 | 22,007 | 字节跳动Seed团队 | Proprietary |
| 45 | Qwen3.5 Max Preview阿里巴巴 | 1514.00 | +/-8 | 6,359 | 阿里巴巴 | Proprietary |
| 46 |  Grok 4.5xAI | 1514.00 | +/-7 | 10,951 | xAI | Proprietary |
| 47 | GPT-5.5 InstantOpenAI | 1513.00 | +/-8 | 7,912 | OpenAI | Proprietary |
| 48 | GLM 5.1智谱AI | 1513.00 | +/-6 | 15,944 | 智谱AI | MIT |
| 49 |  ERNIE-5.1-Preview百度 | 1512.00 | +/-7 | 10,904 | 百度 | Proprietary |
| 50 | Claude Sonnet 4.5Anthropic | 1512.00 | +/-5 | 19,734 | Anthropic | Proprietary |
| 51 | glm-5.2-maxZ.ai | 1512.00 | +/-6 | 12,316 | Z.ai | MIT |
| 52 | Qwen3.6-Max-Preview阿里巴巴 | 1512.00 | +/-15 | 1,606 | 阿里巴巴 | Proprietary |
| 53 | gpt-6-luna-maxOpenAI | 1511.00 | +/-14 | 1,822 | OpenAI | Proprietary |
| 54 | Opus 4.1 (thinking-16k)Anthropic | 1511.00 | +/-6 | 9,878 | Anthropic | Proprietary |
| 55 | GPT-5.4OpenAI | 1511.00 | +/-6 | 18,903 | OpenAI | Proprietary |
| 56 | grok-4.20-beta-0309-reasoningxAI | 1511.00 | +/-6 | 18,334 | xAI | Proprietary |
| 57 | GPT-5.5OpenAI | 1511.00 | +/-6 | 20,019 | OpenAI | Proprietary |
| 58 | grok-4.20-multi-agent-beta-0309xAI | 1509.00 | +/-6 | 18,000 | xAI | Proprietary |
| 59 | Grok 4.20 BetaxAI | 1508.00 | +/-8 | 7,490 | xAI | Proprietary |
| 60 | gemini-3.5-flash-highGoogle | 1507.00 | +/-6 | 13,673 | Proprietary | |
| 61 | grok-4.6-highSpaceXAI | 1507.00 | +/-8 | 5,871 | SpaceXAI | Proprietary |
| 62 | gemini-3.5-flash-mediumGoogle | 1506.00 | +/-6 | 13,480 | Proprietary | |
| 63 | deepseek-v4-pro-high-20260813DeepSeek | 1506.00 | +/-11 | 2,815 | DeepSeek | MIT |
| 64 | Gemini 3.0 FlashGoogle DeepMind | 1505.00 | +/-8 | 6,586 | Google DeepMind | Proprietary |
| 65 | qwen3.8-27bAlibaba | 1505.00 | +/-9 | 5,005 | Alibaba | Apache 2.0 |
| 66 | Opus 4.1Anthropic | 1505.00 | +/-5 | 15,650 | Anthropic | Proprietary |
| 67 | mimo-v2-proXiaomi | 1503.00 | +/-8 | 7,255 | Xiaomi | Proprietary |
| 68 | Kimi K2.5 InstantMoonshot AI | 1503.00 | +/-14 | 1,868 | Moonshot AI | Modified MIT |
| 69 | Kimi K2 ThinkingMoonshot AI | 1502.00 | +/-5 | 20,732 | Moonshot AI | Modified MIT |
| 70 | Gemma 4 31BGoogle DeepMind | 1502.00 | +/-15 | 1,421 | Google DeepMind | Apache 2.0 |
| 71 | GPT-5.6 Luna (xhigh)OpenAI | 1502.00 | +/-7 | 10,293 | OpenAI | Proprietary |
| 72 |  longcat-flash-chat-2602-expMeituan | 1501.00 | +/-7 | 8,146 | Meituan | Proprietary |
| 73 |  Hy3腾讯AI实验室 | 1501.00 | +/-11 | 2,863 | 腾讯AI实验室 | Apache 2.0 |
| 74 | gemini-3.5-flash-liteGoogle | 1501.00 | +/-7 | 9,763 | Proprietary | |
| 75 | Qwen3.7-Plus阿里巴巴 | 1501.00 | +/-6 | 12,160 | 阿里巴巴 | Proprietary |
| 76 | DeepSeek-V4-ProDeepSeek-AI | 1500.00 | +/-6 | 17,135 | DeepSeek-AI | MIT |
| 77 | Grok 4.1 ThinkingxAI | 1500.00 | +/-6 | 15,408 | xAI | Proprietary |
| 78 | GLM-5智谱AI | 1498.00 | +/-7 | 7,621 | 智谱AI | MIT |
| 79 | Claude Opus 4 (thinking-16k)Anthropic | 1498.00 | +/-8 | 6,557 | Anthropic | Proprietary |
| 80 | GPT-5.3 ChatOpenAI | 1498.00 | +/-7 | 9,164 | OpenAI | Proprietary |
| 81 | Qwen 3.6 Plus Preview阿里巴巴 | 1496.00 | +/-6 | 14,664 | 阿里巴巴 | Proprietary |
| 82 |  MiniMax M3MiniMaxAI | 1495.00 | +/-6 | 16,800 | MiniMaxAI | MiniMax Community License |
| 83 | GPT-5.4 mini (high)OpenAI | 1495.00 | +/-6 | 17,208 | OpenAI | Proprietary |
| 84 | Grok 4.1xAI | 1492.00 | +/-6 | 15,992 | xAI | Proprietary |
| 85 | grok-4.7-xhighSpaceXAI | 1492.00 | +/-17 | 1,238 | SpaceXAI | Proprietary |
| 86 | GPT-5.1 Pro (high)OpenAI | 1492.00 | +/-7 | 8,411 | OpenAI | Proprietary |
| 87 | InklingThinking Machines Lab | 1492.00 | +/-7 | 9,634 | Thinking Machines Lab | Apache 2.0 |
| 88 | Qwen3.5-397B-A17B阿里巴巴 | 1491.00 | +/-5 | 25,141 | 阿里巴巴 | Apache 2.0 |
| 89 | mimo-v2.5Xiaomi | 1491.00 | +/-7 | 13,647 | Xiaomi | MIT |
| 90 | ERNIE 5.0 (0110)百度 | 1490.00 | +/-7 | 8,965 | 百度 | Proprietary |
| 91 | Gemini 3.0 Flash (minimal)Google DeepMind | 1490.00 | +/-5 | 23,952 | Google DeepMind | Proprietary |
| 92 | GPT-5.2 Pro (high)OpenAI | 1489.00 | +/-6 | 12,226 | OpenAI | Proprietary |
| 93 | mimo-v2-omniXiaomi | 1489.00 | +/-9 | 5,878 | Xiaomi | Proprietary |
| 94 | GLM-5V-Turbo智谱AI | 1488.00 | +/-11 | 2,918 | 智谱AI | Proprietary |
| 95 | deepseek-v4-pro-high-previewDeepSeek | 1488.00 | +/-6 | 16,000 | DeepSeek | MIT |
| 96 | Grok 4.3 BetaxAI | 1488.00 | +/-6 | 19,933 | xAI | Proprietary |
| 97 | Kimi K2 Thinking (thinking-turbo)Moonshot AI | 1486.00 | +/-5 | 15,272 | Moonshot AI | Modified MIT |
| 98 |  amazon-nova-experimental-chat-26-02-10Amazon | 1486.00 | +/-19 | 879 | Amazon | Proprietary |
| 99 | GLM-4.7智谱AI | 1484.00 | +/-12 | 2,485 | 智谱AI | MIT |
| 100 | DeepSeek-V4-FlashDeepSeek-AI | 1484.00 | +/-6 | 15,253 | DeepSeek-AI | MIT |
| 101 | Gemma 4 26B A4BGoogle DeepMind | 1483.00 | +/-15 | 1,443 | Google DeepMind | Apache 2.0 |
| 102 | GPT-5.2OpenAI | 1482.00 | +/-5 | 22,091 | OpenAI | Proprietary |
| 103 | Haiku 4.5Anthropic | 1481.00 | +/-4 | 36,871 | Anthropic | Proprietary |
| 104 | amazon-nova-experimental-chat-26-01-10Amazon | 1480.00 | +/-21 | 781 | Amazon | Proprietary |
| 105 | Qwen3 Max (Preview)阿里巴巴 | 1480.00 | +/-8 | 5,336 | 阿里巴巴 | Proprietary |
| 106 | MiniMax-M2.7MiniMaxAI | 1480.00 | +/-5 | 22,747 | MiniMaxAI | Modified MIT |
| 107 | Muse Glimmer-30BFacebook AI研究实验室 | 1480.00 | +/-18 | 1,119 | Facebook AI研究实验室 | Apache-2.0 |
| 108 | deepseek-v4-flash-high-previewDeepSeek | 1479.00 | +/-6 | 15,210 | DeepSeek | MIT |
| 109 |  Mistral Medium 3.5MistralAI | 1479.00 | +/-11 | 3,281 | MistralAI | Modified MIT |
| 110 | GPT-5.1OpenAI | 1476.00 | +/-7 | 9,341 | OpenAI | Proprietary |
| 111 | DeepSeek V3.2-Exp (thinking)DeepSeek-AI | 1476.00 | +/-7 | 8,876 | DeepSeek-AI | MIT |
| 112 | DeepSeek V3.2-Exp (thinking)DeepSeek-AI | 1475.00 | +/-13 | 1,916 | DeepSeek-AI | MIT |
| 113 |  Nemotron 3 UltraNVIDIA | 1474.00 | +/-11 | 3,195 | NVIDIA | OpenMDW-1.1 |
| 114 | Claude Sonnet 4 (thinking-32k)Anthropic | 1474.00 | +/-8 | 6,312 | Anthropic | Proprietary |
| 115 | Inkling SmallThinky | 1474.00 | +/-8 | 7,383 | Thinky | Apache 2.0 |
| 116 | qwen3-max-2025-09-23Alibaba | 1473.00 | +/-13 | 2,017 | Alibaba | Proprietary |
| 117 | Qwen3-235B-A22B-2507阿里巴巴 | 1472.00 | +/-5 | 21,697 | 阿里巴巴 | Apache 2.0 |
| 118 | longcat-flash-chatMeituan | 1472.00 | +/-13 | 2,203 | Meituan | MIT |
| 119 | ERNIE 5.0 Preview (1203)百度 | 1472.00 | +/-13 | 1,992 | 百度 | Proprietary |
| 120 | GPT-5-Pro (high)OpenAI | 1470.00 | +/-8 | 6,352 | OpenAI | Proprietary |
| 121 | DeepSeek V3.2DeepSeek-AI | 1469.00 | +/-6 | 10,909 | DeepSeek-AI | MIT |
| 122 | Mistral Large 3MistralAI | 1469.00 | +/-5 | 19,911 | MistralAI | Apache 2.0 |
| 123 | Kimi K2 0905Moonshot AI | 1467.00 | +/-13 | 2,224 | Moonshot AI | Modified MIT |
| 124 | GPT-4o(2025-03-27)OpenAI | 1467.00 | +/-5 | 15,818 | OpenAI | Proprietary |
| 125 | DeepSeek V3.2-ExpDeepSeek-AI | 1467.00 | +/-12 | 2,537 | DeepSeek-AI | MIT |
| 126 | Claude Opus 4Anthropic | 1465.00 | +/-7 | 7,791 | Anthropic | Proprietary |
| 127 | Gemini 2.5 ProGoogle DeepMind | 1464.00 | +/-4 | 26,883 | Google DeepMind | Proprietary |
| 128 | DeepSeek-R1-0528DeepSeek-AI | 1464.00 | +/-11 | 2,681 | DeepSeek-AI | MIT |
| 129 | solar-pro4Upstage | 1463.00 | +/-10 | 4,073 | Upstage | Proprietary |
| 130 | Qwen3-VL-235B-A22B-Instruct阿里巴巴 | 1463.00 | +/-13 | 2,291 | 阿里巴巴 | Apache 2.0 |
| 131 | GPT-5.2 Chat (0210)OpenAI | 1462.00 | +/-8 | 5,952 | OpenAI | Proprietary |
| 132 | Grok 4.1 Fast (fast-reasoning)xAI | 1461.00 | +/-6 | 13,923 | xAI | Proprietary |
| 133 | DeepSeek-V3.1 Terminus (thinking)DeepSeek-AI | 1460.00 | +/-24 | 631 | DeepSeek-AI | MIT |
| 134 | OpenAI o3OpenAI | 1460.00 | +/-6 | 11,537 | OpenAI | Proprietary |
| 135 | GPT-5.4 nano (high)OpenAI | 1460.00 | +/-6 | 17,640 | OpenAI | Proprietary |
| 136 | Kimi K2Moonshot AI | 1460.00 | +/-8 | 5,169 | Moonshot AI | Modified MIT |
| 137 | Qwen3.5-122B-A10B阿里巴巴 | 1458.00 | +/-7 | 8,172 | 阿里巴巴 | Apache 2.0 |
| 138 | GPT-4.5OpenAI | 1458.00 | +/-13 | 1,939 | OpenAI | Proprietary |
| 139 | Grok 4 FastxAI | 1458.00 | +/-17 | 1,181 | xAI | Proprietary |
| 140 | GLM-4.6智谱AI | 1458.00 | +/-7 | 7,572 | 智谱AI | MIT |
| 141 | Gemini 3.1 Flash-LiteGoogle DeepMind | 1458.00 | +/-6 | 17,816 | Google DeepMind | Proprietary |
| 142 | Qwen3-Coder-480B-A35B阿里巴巴 | 1457.00 | +/-9 | 4,774 | 阿里巴巴 | Apache 2.0 |
| 143 | GPT-4.1OpenAI | 1456.00 | +/-7 | 9,191 | OpenAI | Proprietary |
| 144 | DeepSeek-V3.1 (thinking)DeepSeek-AI | 1456.00 | +/-13 | 1,871 | DeepSeek-AI | MIT |
| 145 | GLM-4.5智谱AI | 1455.00 | +/-9 | 4,711 | 智谱AI | MIT |
| 146 | Magistral-Medium-2506MistralAI | 1454.00 | +/-5 | 21,502 | MistralAI | Proprietary |
| 147 | hunyuan-hy3-previewTencent | 1453.00 | +/-13 | 2,052 | Tencent | tencent-hunyuan-community |
| 148 | Claude3-Sonnet (thinking-32k)Anthropic | 1453.00 | +/-8 | 6,120 | Anthropic | Proprietary |
| 149 | Qwen3.5-27B阿里巴巴 | 1452.00 | +/-7 | 7,937 | 阿里巴巴 | Apache 2.0 |
| 150 | Qwen3-VL-235B-A22B-Instruct (thinking)阿里巴巴 | 1451.00 | +/-14 | 1,626 | 阿里巴巴 | Apache 2.0 |
| 151 |  Step 3.5 FlashStepFunAI | 1449.00 | +/-6 | 15,171 | StepFunAI | Apache 2.0 |
| 152 | Claude Sonnet 4Anthropic | 1449.00 | +/-7 | 7,265 | Anthropic | Proprietary |
| 153 | DeepSeek-V3.1DeepSeek-AI | 1447.00 | +/-12 | 2,594 | DeepSeek-AI | MIT |
| 154 | mimo-v2-flash (non-thinking)Xiaomi | 1447.00 | +/-6 | 12,452 | Xiaomi | MIT |
| 155 | MiniMax M2.5MiniMaxAI | 1446.00 | +/-6 | 11,348 | MiniMaxAI | Modified MIT |
| 156 | qwen3-235b-a22b-no-thinkingAlibaba | 1445.00 | +/-8 | 6,893 | Alibaba | Apache 2.0 |
| 157 | DeepSeek-R1DeepSeek-AI | 1444.00 | +/-12 | 2,317 | DeepSeek-AI | MIT |
| 158 | Qwen3-Next阿里巴巴 | 1444.00 | +/-9 | 4,779 | 阿里巴巴 | Apache 2.0 |
| 159 | trinity-large-preview Apache 2.0 | 1443.00 | +/-7 | 7,972 | — | — |
| 160 | Grok 3xAI | 1442.00 | +/-8 | 5,345 | xAI | Proprietary |
| 161 | qwen3-235b-a22b-thinking-2507Alibaba | 1442.00 | +/-15 | 1,587 | Alibaba | Apache 2.0 |
| 162 | DeepSeek-V3.1 TerminusDeepSeek-AI | 1440.00 | +/-21 | 770 | DeepSeek-AI | MIT |
| 163 | M2.1MiniMaxAI | 1438.00 | +/-10 | 3,483 | MiniMaxAI | MIT |
| 164 | Grok 4 Fast (fast-reasoning)xAI | 1438.00 | +/-9 | 3,901 | xAI | Proprietary |
| 165 | Qwen3-30B-A3B-2507阿里巴巴 | 1438.00 | +/-9 | 4,594 | 阿里巴巴 | Apache 2.0 |
| 166 | hunyuan-vision-1.5-thinkingTencent | 1436.00 | +/-27 | 419 | Tencent | Proprietary |
| 167 | Claude 3.5 SonnetAnthropic | 1436.00 | +/-6 | 14,898 | Anthropic | Proprietary |
| 168 | Step 3.5 FlashStepFunAI | 1436.00 | +/-6 | 17,684 | StepFunAI | Proprietary |
| 169 | Qwen3.5-35B-A3B阿里巴巴 | 1435.00 | +/-7 | 8,392 | 阿里巴巴 | Apache 2.0 |
| 170 | Grok 4xAI | 1435.00 | +/-7 | 7,984 | xAI | Proprietary |
| 171 | OpenAI o3-mini (high)OpenAI | 1434.00 | +/-12 | 2,596 | OpenAI | Proprietary |
| 172 | amazon-nova-experimental-chat-12-10Amazon | 1434.00 | +/-21 | 725 | Amazon | Proprietary |
| 173 | OpenAI o4 - miniOpenAI | 1433.00 | +/-7 | 8,613 | OpenAI | Proprietary |
| 174 | GPT-4.1 miniOpenAI | 1433.00 | +/-8 | 6,801 | OpenAI | Proprietary |
| 175 | mimo-v2-flash (thinking)Xiaomi | 1433.00 | +/-12 | 2,491 | Xiaomi | MIT |
| 176 | OpenAI o1OpenAI | 1433.00 | +/-10 | 3,973 | OpenAI | Proprietary |
| 177 | mistral-medium-2505Mistral | 1433.00 | +/-8 | 5,826 | Mistral | Proprietary |
| 178 | Qwen3-235B-A22B阿里巴巴 | 1432.00 | +/-9 | 4,302 | 阿里巴巴 | Apache 2.0 |
| 179 | ERNIE 5.0百度 | 1432.00 | +/-19 | 925 | 百度 | Proprietary |
| 180 | Claude3-SonnetAnthropic | 1431.00 | +/-7 | 7,054 | Anthropic | Proprietary |
| 181 | GPT-5-mini (high)OpenAI | 1431.00 | +/-8 | 5,481 | OpenAI | Proprietary |
| 182 | DeepSeek-V3-0324DeepSeek-AI | 1428.00 | +/-7 | 8,292 | DeepSeek-AI | MIT |
| 183 | Gemini 2.5 Flash-Preview-09-2025Google DeepMind | 1427.00 | +/-7 | 6,884 | Google DeepMind | Proprietary |
| 184 | GLM-4.5-Air智谱AI | 1427.00 | +/-8 | 6,057 | 智谱AI | MIT |
| 185 | Gemini 2.5 FlashGoogle DeepMind | 1424.00 | +/-4 | 26,315 | Google DeepMind | Proprietary |
| 186 | GLM-4.7-Flash智谱AI | 1423.00 | +/-11 | 2,787 | 智谱AI | MIT |
| 187 | Qwen3-Next (thinking)阿里巴巴 | 1421.00 | +/-11 | 2,645 | 阿里巴巴 | Apache 2.0 |
| 188 | nvidia-nemotron-3.5-lightning-30b-a3b-nvfp4Nvidia | 1420.00 | +/-10 | 3,691 | Nvidia | OpenMDW-1.1 |
| 189 | amazon-nova-experimental-chat-11-10Amazon | 1420.00 | +/-8 | 5,522 | Amazon | Proprietary |
| 190 | OpenAI o3-miniOpenAI | 1417.00 | +/-6 | 9,380 | OpenAI | Proprietary |
| 191 | minimax-m1MiniMax | 1416.00 | +/-8 | 6,357 | MiniMax | Apache 2.0 |
| 192 | GLM-4.6V智谱AI | 1416.00 | +/-25 | 545 | 智谱AI | MIT |
| 193 | intellect-3 MIT | 1416.00 | +/-18 | 1,002 | — | — |
| 194 | OpenAI o1OpenAI | 1416.00 | +/-9 | 5,123 | OpenAI | Proprietary |
| 195 | trinity-large-thinking Apache 2.0 | 1413.00 | +/-7 | 8,542 | — | — |
| 196 | Mistral-Small-3.2MistralAI | 1412.00 | +/-10 | 3,326 | MistralAI | Apache 2.0 |
| 197 | amazon-nova-experimental-chat-10-20Amazon | 1410.00 | +/-12 | 2,329 | Amazon | Proprietary |
| 198 | ling-flash-2.0 AntGroup | 1410.00 | +/-15 | 1,527 | Group | MIT |
| 199 | Step3StepFunAI | 1409.00 | +/-17 | 1,207 | StepFunAI | Apache 2.0 |
| 200 | nvidia-nemotron-3-super-120b-a12bNvidia | 1407.00 | +/-13 | 1,884 | Nvidia | NVIDIA Open Model |
| 201 | Qwen3-32B阿里巴巴 | 1406.00 | +/-24 | 513 | 阿里巴巴 | Apache 2.0 |
| 202 | GLM-4.5V智谱AI | 1403.00 | +/-19 | 966 | 智谱AI | MIT |
| 203 | nvidia-llama-3.3-nemotron-super-49b-v1.5Nvidia | 1402.00 | +/-22 | 651 | Nvidia | Nvidia Open |
| 204 | Qwen2.5-Max阿里巴巴 | 1402.00 | +/-8 | 5,074 | 阿里巴巴 | Proprietary |
| 205 | Hunyuan-T1腾讯AI实验室 | 1401.00 | +/-20 | 798 | 腾讯AI实验室 | Proprietary |
| 206 | Claude 3.5 SonnetAnthropic | 1399.00 | +/-7 | 13,607 | Anthropic | Proprietary |
| 207 | hunyuan-turbos-20250226Tencent | 1399.00 | +/-30 | 275 | Tencent | Proprietary |
| 208 | granite-4.2-30bIBM | 1399.00 | +/-21 | 857 | IBM | Apache 2.0 |
| 209 | Gemini 2.5 Flash-Lite-Preview-09-2025 (no-thinking)Google DeepMind | 1397.00 | +/-6 | 9,824 | Google DeepMind | Proprietary |
| 210 | hunyuan-turbos-20250416Tencent | 1394.00 | +/-14 | 1,757 | Tencent | Proprietary |
| 211 | Nova 2 Lite亚马逊 | 1393.00 | +/-12 | 2,578 | 亚马逊 | Proprietary |
| 212 | OpenAI o3-mini (high)OpenAI | 1393.00 | +/-10 | 3,150 | OpenAI | Proprietary |
| 213 | amazon-nova-experimental-chat-10-09Amazon | 1392.00 | +/-24 | 542 | Amazon | Proprietary |
| 214 | mercury-2 InceptionAI | 1392.00 | +/-20 | 804 | AI | Proprietary |
| 215 | llama-3.1-nemotron-ultra-253b-v1Nvidia | 1391.00 | +/-30 | 367 | Nvidia | Nvidia Open Model |
| 216 | ring-flash-2.0 AntGroup | 1390.00 | +/-15 | 1,492 | Group | MIT |
| 217 | GPT OSS 120BOpenAI | 1390.00 | +/-8 | 6,374 | OpenAI | Apache 2.0 |
| 218 |  C4AI Command A (202503)CohereAI | 1389.00 | +/-6 | 10,162 | CohereAI | CC-BY-NC-4.0 |
| 219 | Magistral-Medium-2506MistralAI | 1387.00 | +/-12 | 2,232 | MistralAI | Proprietary |
| 220 | DeepSeek-V3DeepSeek-AI | 1387.00 | +/-10 | 3,280 | DeepSeek-AI | DeepSeek |
| 221 | OpenAI o1-miniOpenAI | 1386.00 | +/-7 | 8,478 | OpenAI | Proprietary |
| 222 | grok-3-mini-betaxAI | 1386.00 | +/-9 | 4,081 | xAI | Proprietary |
| 223 | Claude 3.5 HaikuAnthropic | 1386.00 | +/-6 | 11,120 | Anthropic | Proprietary |
| 224 | Qwen3-30B-A3B阿里巴巴 | 1385.00 | +/-9 | 4,486 | 阿里巴巴 | Apache 2.0 |
| 225 | GPT-5-Nano (high)OpenAI | 1384.00 | +/-15 | 1,663 | OpenAI | Proprietary |
| 226 | QwQ-32B阿里巴巴 | 1383.00 | +/-9 | 4,007 | 阿里巴巴 | Apache 2.0 |
| 227 | Gemini 2.5 Flash-Lite (thinking)Google DeepMind | 1382.00 | +/-8 | 5,943 | Google DeepMind | Proprietary |
| 228 | MiniMax M2MiniMaxAI | 1381.00 | +/-15 | 1,537 | MiniMaxAI | Apache 2.0 |
| 229 | olmo-3.1-32b-instructAi2 | 1381.00 | +/-12 | 2,467 | Ai2 | Apache 2.0 |
| 230 | granite-4.2-3bIBM | 1379.00 | +/-22 | 838 | IBM | Apache 2.0 |
| 231 | qwen-plus-0125Alibaba | 1379.00 | +/-18 | 893 | Alibaba | Proprietary |
| 232 | llama-3.1-405b-instruct-bf16Meta | 1376.00 | +/-8 | 6,249 | Meta | Llama 3.1 Community |
| 233 | deepseek-v2.5-1210DeepSeek | 1374.00 | +/-17 | 1,079 | DeepSeek | DeepSeek |
| 234 | GPT-4.1 nanoOpenAI | 1373.00 | +/-19 | 807 | OpenAI | Proprietary |
| 235 | Llama 4 Maverick InstructFacebook AI研究实验室 | 1373.00 | +/-8 | 6,917 | Facebook AI研究实验室 | Llama 4 |
| 236 | step-2-16k-exp-202412StepFun | 1371.00 | +/-20 | 737 | StepFun | Proprietary |
| 237 | hunyuan-turbo-0110Tencent | 1371.00 | +/-30 | 299 | Tencent | Proprietary |
| 238 | granite-4.2-8bIBM | 1369.00 | +/-21 | 836 | IBM | Apache 2.0 |
| 239 | GPT-4oOpenAI | 1369.00 | +/-6 | 19,526 | OpenAI | Proprietary |
| 240 | llama-3.1-405b-instruct-fp8Meta | 1369.00 | +/-7 | 9,714 | Meta | Llama 3.1 Community |
| 241 | athene-v2-chat NexusFlow | 1369.00 | +/-9 | 4,019 | — | — |
| 242 | GPT OSS 20BOpenAI | 1369.00 | +/-13 | 2,105 | OpenAI | Apache 2.0 |
| 243 | yi-lightning Proprietary | 1369.00 | +/-10 | 4,316 | — | — |
| 244 | DeepSeek V2.5DeepSeek-AI | 1368.00 | +/-9 | 4,252 | DeepSeek-AI | DeepSeek |
| 245 | hunyuan-large-2025-02-10Tencent | 1366.00 | +/-25 | 519 | Tencent | Proprietary |
| 246 | mercury InceptionAI | 1364.00 | +/-28 | 408 | AI | Proprietary |
| 247 | Gemini 2.0 Flash ExperimentalGoogle DeepMind | 1364.00 | +/-7 | 6,937 | Google DeepMind | Proprietary |
| 248 | Llama 4 Scout InstructFacebook AI研究实验室 | 1362.00 | +/-9 | 5,204 | Facebook AI研究实验室 | Llama |
| 249 | llama-3.3-nemotron-49b-super-v1Nvidia | 1362.00 | +/-31 | 286 | Nvidia | Nvidia |
| 250 | Mistral-Small-3.1-24B-Instruct-2503MistralAI | 1361.00 | +/-8 | 6,084 | MistralAI | Apache 2.0 |
| 251 | nvidia-nemotron-3-nano-30b-a3b-bf16Nvidia | 1361.00 | +/-10 | 3,328 | Nvidia | NVIDIA Open Model |
| 252 | olmo-3-32b-thinkAi2 | 1360.00 | +/-18 | 1,040 | Ai2 | Apache 2.0 |
| 253 | GPT-4oOpenAI | 1360.00 | +/-8 | 7,318 | OpenAI | Proprietary |
| 254 | Grok 2xAI | 1358.00 | +/-7 | 10,368 | xAI | Proprietary |
| 255 | Gemma 3 - 27B (IT)Google DeepMind | 1358.00 | +/-7 | 7,982 | Google DeepMind | Gemma |
| 256 | hunyuan-large-visionTencent | 1357.00 | +/-19 | 951 | Tencent | Proprietary |
| 257 | qwen2.5-plus-1127Alibaba | 1356.00 | +/-14 | 1,553 | Alibaba | Proprietary |
| 258 | Gemini 1.5 ProGoogle DeepMind | 1356.00 | +/-7 | 9,175 | Google DeepMind | Proprietary |
| 259 | Claude3-OpusAnthropic | 1355.00 | +/-6 | 33,748 | Anthropic | Proprietary |
| 260 | Qwen2.5-VL-72B-Instruct阿里巴巴 | 1355.00 | +/-8 | 6,688 | 阿里巴巴 | Qwen |
| 261 | mistral-large-2407Mistral | 1354.00 | +/-8 | 7,589 | Mistral | Mistral Research |
| 262 | granite-4.1-8bIBM | 1354.00 | +/-20 | 1,105 | IBM | Apache 2.0 |
| 263 | qwen-max-0919Alibaba | 1353.00 | +/-11 | 2,756 | Alibaba | Qwen |
| 264 | glm-4-plusZhipu AI | 1352.00 | +/-9 | 4,449 | Zhipu AI | Proprietary |
| 265 | step-1o-turbo-202506StepFun | 1351.00 | +/-15 | 1,482 | StepFun | Proprietary |
| 266 | athene-70b-0725 CC-BY-NC-4.0 | 1350.00 | +/-11 | 3,122 | — | — |
| 267 | GPT-4o miniOpenAI | 1348.00 | +/-7 | 10,926 | OpenAI | Proprietary |
| 268 | gpt-4-turbo-2024-04-09OpenAI | 1347.00 | +/-7 | 17,104 | OpenAI | Proprietary |
| 269 | Gemini 1.5 ProGoogle DeepMind | 1346.00 | +/-8 | 12,747 | Google DeepMind | Proprietary |
| 270 | mistral-large-2411Mistral | 1346.00 | +/-9 | 4,212 | Mistral | MRL |
| 271 | Llama3.3-70B-InstructFacebook AI研究实验室 | 1346.00 | +/-7 | 8,717 | Facebook AI研究实验室 | Llama-3.3 |
| 272 | amazon-nova-pro-v1.0Amazon | 1343.00 | +/-9 | 3,853 | Amazon | Proprietary |
| 273 | deepseek-coder-v2DeepSeek | 1342.00 | +/-12 | 2,671 | DeepSeek | DeepSeek License |
| 274 | Gemini 2.0 Flash-LiteGoogle DeepMind | 1342.00 | +/-10 | 3,474 | Google DeepMind | Proprietary |
| 275 | Qwen2.5-Coder-32B-Instruct阿里巴巴 | 1342.00 | +/-18 | 873 | 阿里巴巴 | Apache 2.0 |
| 276 | olmo-3.1-32b-thinkAi2 | 1341.00 | +/-15 | 1,554 | Ai2 | Apache 2.0 |
| 277 | GPT-4OpenAI | 1340.00 | +/-7 | 15,605 | OpenAI | Proprietary |
| 278 | gemini-advanced-0514Google | 1337.00 | +/-9 | 8,138 | Proprietary | |
| 279 | grok-2-mini-2024-08-13xAI | 1335.00 | +/-7 | 8,652 | xAI | Proprietary |
| 280 | Llama3.1-70B-InstructFacebook AI研究实验室 | 1333.00 | +/-7 | 9,389 | Facebook AI研究实验室 | Llama 3.1 Community |
| 281 | GPT-4OpenAI | 1332.00 | +/-8 | 15,289 | OpenAI | Proprietary |
| 282 | hunyuan-standard-2025-02-10Tencent | 1331.00 | +/-24 | 549 | Tencent | Proprietary |
| 283 | glm-4-plus-0111Zhipu | 1330.00 | +/-18 | 894 | Zhipu | Proprietary |
| 284 | GPT-4OpenAI | 1330.00 | +/-9 | 8,306 | OpenAI | Proprietary |
| 285 | Llama3.1-70B-InstructFacebook AI研究实验室 | 1328.00 | +/-15 | 1,312 | Facebook AI研究实验室 | Llama 3.1 |
| 286 | ibm-granite-h-smallIBM | 1328.00 | +/-17 | 1,275 | IBM | Apache 2.0 |
| 287 | Claude3-SonnetAnthropic | 1318.00 | +/-7 | 18,888 | Anthropic | Proprietary |
| 288 | gemini-1.5-flash-002Google | 1316.00 | +/-8 | 5,892 | Proprietary | |
| 289 | Gemma 3 - 12B (IT)Google DeepMind | 1316.00 | +/-23 | 543 | Google DeepMind | Gemma |
| 290 | reka-core-20240904 Proprietary | 1315.00 | +/-15 | 1,216 | — | — |
| 291 | GPT-4OpenAI | 1315.00 | +/-8 | 13,719 | OpenAI | Proprietary |
| 292 | llama-3.1-nemotron-51b-instructNvidia | 1312.00 | +/-22 | 665 | Nvidia | Llama 3.1 |
| 293 | Mistral Small 24B Instruct 2501MistralAI | 1312.00 | +/-12 | 2,083 | MistralAI | Apache 2.0 |
| 294 | jamba-1.5-large Jamba Open | 1312.00 | +/-15 | 1,440 | — | — |
| 295 | GLM4智谱AI | 1309.00 | +/-14 | 1,718 | 智谱AI | Proprietary |
| 296 | gemini-1.5-flash-001Google | 1309.00 | +/-8 | 10,680 | Proprietary | |
| 297 | nemotron-4-340b-instructNvidia | 1309.00 | +/-11 | 3,254 | Nvidia | NVIDIA Open Model |
| 298 | llama-3.1-tulu-3-70bAi2 | 1308.00 | +/-24 | 450 | Ai2 | Llama 3.1 |
| 299 | Gemma-3n-E4BGoogle DeepMind | 1307.00 | +/-10 | 3,481 | Google DeepMind | Gemma |
| 300 | Llama3-70B-InstructFacebook AI研究实验室 | 1306.00 | +/-7 | 28,126 | Facebook AI研究实验室 | Llama 3 Community |
| 301 |  Phi-4-reasoningMicrosoft Azure | 1306.00 | +/-10 | 3,305 | Microsoft Azure | MIT |
| 302 | amazon-nova-lite-v1.0Amazon | 1306.00 | +/-10 | 3,060 | Amazon | Proprietary |
| 303 | gemma-2-27b-itGoogle | 1304.00 | +/-6 | 12,088 | Gemma license | |
| 304 | Claude3-HaikuAnthropic | 1301.00 | +/-7 | 20,898 | Anthropic | Proprietary |
| 305 | hunyuan-standard-256kTencent | 1301.00 | +/-24 | 497 | Tencent | Proprietary |
| 306 | Qwen2-72B-Instruct阿里巴巴 | 1297.00 | +/-9 | 6,249 | 阿里巴巴 | Qianwen LICENSE |
| 307 | mistral-large-2402Mistral | 1296.00 | +/-9 | 10,418 | Mistral | Proprietary |
| 308 | C4AI Aya Vision 32BCohereAI | 1292.00 | +/-9 | 4,685 | CohereAI | CC-BY-NC-4.0 |
| 309 | reka-flash-20240904 Proprietary | 1292.00 | +/-15 | 1,207 | — | — |
| 310 | amazon-nova-micro-v1.0Amazon | 1288.00 | +/-10 | 2,981 | Amazon | Proprietary |
| 311 | Llama3.1-8B-InstructFacebook AI研究实验室 | 1288.00 | +/-26 | 478 | Facebook AI研究实验室 | Apache 2.0 |
| 312 | command-r-08-2024Cohere | 1282.00 | +/-13 | 1,783 | Cohere | CC-BY-NC-4.0 |
| 313 | command-r-plus-08-2024Cohere | 1280.00 | +/-14 | 1,675 | Cohere | CC-BY-NC-4.0 |
| 314 | Qwen1.5-110B-Chat阿里巴巴 | 1280.00 | +/-10 | 4,763 | 阿里巴巴 | Qianwen LICENSE |
| 315 | olmo-2-0325-32b-instructAi2 | 1280.00 | +/-27 | 427 | Ai2 | Apache-2.0 |
| 316 | reka-flash-21b-20240226-online Proprietary | 1278.00 | +/-13 | 2,879 | — | — |
| 317 | Mixtral-8x22B-Instruct-v0.1MistralAI | 1278.00 | +/-9 | 8,780 | MistralAI | Apache 2.0 |
| 318 | Qwen1.5-72B-Chat阿里巴巴 | 1275.00 | +/-10 | 6,370 | 阿里巴巴 | Qianwen LICENSE |
| 319 | gpt-3.5-turbo-0125OpenAI | 1275.00 | +/-8 | 11,130 | OpenAI | Proprietary |
| 320 | ministral-8b-2410Mistral | 1274.00 | +/-19 | 838 | Mistral | MRL |
| 321 | Gemma 3 - 4B (IT)Google DeepMind | 1274.00 | +/-24 | 605 | Google DeepMind | Gemma |
| 322 | C4AI Command R+CohereAI | 1273.00 | +/-8 | 13,937 | CohereAI | CC-BY-NC-4.0 |
| 323 | gemini-1.5-flash-8b-001Google | 1271.00 | +/-8 | 6,069 | Proprietary | |
| 324 | gemma-2-9b-it-simpo MIT | 1271.00 | +/-15 | 1,471 | — | — |
| 325 | gemma-2-9b-itGoogle | 1270.00 | +/-7 | 8,921 | Gemma license | |
| 326 | reka-flash-21b-20240226 Proprietary | 1267.00 | +/-11 | 4,748 | — | — |
| 327 | jamba-1.5-mini Jamba Open | 1265.00 | +/-15 | 1,352 | — | — |
| 328 | mistral-mediumMistral | 1263.00 | +/-10 | 5,149 | Mistral | Proprietary |
| 329 | gpt-3.5-turbo-1106OpenAI | 1263.00 | +/-15 | 2,121 | OpenAI | Proprietary |
| 330 | qwen1.5-32b-chatAlibaba | 1262.00 | +/-11 | 3,930 | Alibaba | Qianwen LICENSE |
| 331 | Llama3.1-8B-InstructFacebook AI研究实验室 | 1260.00 | +/-7 | 8,582 | Facebook AI研究实验室 | Llama 3.1 Community |
| 332 | C4AI Aya Vision 8BCohereAI | 1255.00 | +/-15 | 1,567 | CohereAI | CC-BY-NC-4.0 |
| 333 | Llama3-8B-InstructFacebook AI研究实验室 | 1253.00 | +/-8 | 18,374 | Facebook AI研究实验室 | Llama 3 Community |
| 334 | llama-3.1-tulu-3-8bAi2 | 1253.00 | +/-25 | 476 | Ai2 | Llama 3.1 |
| 335 | dbrx-instruct-preview DBRX LICENSE | 1252.00 | +/-11 | 5,502 | — | — |
| 336 | granite-3.1-2b-instructIBM | 1249.00 | +/-25 | 508 | IBM | Apache 2.0 |
| 337 | Gemini-proGoogle DeepMind | 1249.00 | +/-24 | 678 | Google DeepMind | Proprietary |
| 338 |  InternLM2-Base-20B上海人工智能实验室 | 1249.00 | +/-14 | 1,684 | 上海人工智能实验室 | — |
| 339 |  Yi-1.5-34B零一万物 | 1249.00 | +/-10 | 3,841 | 零一万物 | — |
| 340 | zephyr-orpo-141b-A35b-v0.1 Apache 2.0 | 1245.00 | +/-21 | 831 | — | — |
| 341 | command-rCohere | 1244.00 | +/-9 | 9,645 | Cohere | CC-BY-NC-4.0 |
| 342 | granite-3.0-8b-instructIBM | 1241.00 | +/-18 | 1,108 | IBM | Apache 2.0 |
| 343 | Qwen1.5-14B-Chat阿里巴巴 | 1240.00 | +/-13 | 3,208 | 阿里巴巴 | Qianwen LICENSE |
| 344 | mixtral-8x7b-instruct-v0.1Mistral | 1240.00 | +/-8 | 11,784 | Mistral | Apache 2.0 |
| 345 | gemini-pro-dev-apiGoogle | 1240.00 | +/-14 | 2,681 | Proprietary | |
| 346 | starling-lm-7b-beta Apache-2.0 | 1236.00 | +/-13 | 2,948 | — | — |
| 347 | Phi-3-medium 14B-previewMicrosoft Azure | 1232.00 | +/-10 | 3,973 | Microsoft Azure | MIT |
| 348 | openchat-3.5-0106 Apache-2.0 | 1230.00 | +/-14 | 2,005 | — | — |
| 349 | snowflake-arctic-instruct Apache 2.0 | 1225.00 | +/-11 | 5,734 | — | — |
| 350 | DeepSeek LLM 67B ChatDeepSeek-AI | 1219.00 | +/-23 | 649 | DeepSeek-AI | DeepSeek License |
| 351 |  Gemma 1.1-7B-ITGoogle Research | 1216.00 | +/-10 | 4,332 | Google Research | Gemma license |
| 352 | LLaMA2 70BFacebook AI研究实验室 | 1215.00 | +/-21 | 805 | Facebook AI研究实验室 | — |
| 353 | Qwen3-VL-2B阿里巴巴 | 1210.00 | +/-17 | 1,134 | 阿里巴巴 | Apache 2.0 |
| 354 | Qwen1.5-7B-Chat阿里巴巴 | 1210.00 | +/-21 | 772 | 阿里巴巴 | Qianwen LICENSE |
| 355 | starling-lm-7b-alpha CC-BY-NC-4.0 | 1208.00 | +/-16 | 1,397 | — | — |
| 356 | Yi-34B零一万物 | 1206.00 | +/-13 | 2,345 | 零一万物 | — |
| 357 | Phi-3-small 7BMicrosoft Azure | 1205.00 | +/-12 | 3,219 | Microsoft Azure | MIT |
| 358 | openchat-3.5 Apache-2.0 | 1203.00 | +/-20 | 971 | — | — |
| 359 | Qwen-14B-Chat阿里巴巴 | 1198.00 | +/-23 | 599 | 阿里巴巴 | Qianwen LICENSE |
| 360 | Phi-3-mini 3.8BMicrosoft Azure | 1197.00 | +/-14 | 1,841 | Microsoft Azure | MIT |
| 361 |  Vicuna 33BLM-SYS | 1194.00 | +/-13 | 2,866 | LM-SYS | — |
| 362 |  WizardLM-70B-V1.0WizardLM Team | 1194.00 | +/-20 | 988 | WizardLM Team | Llama 2 Community |
| 363 | gemma-2-2b-itGoogle | 1192.00 | +/-8 | 7,298 | Gemma license | |
| 364 | Phi-3-mini 3.8BMicrosoft Azure | 1188.00 | +/-12 | 3,449 | Microsoft Azure | MIT |
| 365 | openhermes-2.5-mistral-7b Apache-2.0 | 1187.00 | +/-23 | 589 | — | — |
| 366 | Mistral-7B-Instruct-v0.2MistralAI | 1186.00 | +/-12 | 3,114 | MistralAI | Apache-2.0 |
| 367 | solar-10.7b-instruct-v1.0 CC-BY-NC-4.0 | 1184.00 | +/-27 | 482 | — | — |
| 368 | llama-2-70b-chatMeta | 1179.00 | +/-10 | 5,717 | Meta | Llama 2 Community |
| 369 | llama-3.2-3b-instructMeta | 1177.00 | +/-16 | 1,351 | Meta | Llama 3.2 |
| 370 | nous-hermes-2-mixtral-8x7b-dpo Apache-2.0 | 1176.00 | +/-23 | 575 | — | — |
| 371 | QwQ-32B-Preview阿里巴巴 | 1174.00 | +/-24 | 566 | 阿里巴巴 | Apache 2.0 |
| 372 | Gemma 1.1-2B-ITGoogle Research | 1171.00 | +/-14 | 1,963 | Google Research | Gemma license |
| 373 | mpt-30b-chat CC-BY-NC-SA-4.0 | 1168.00 | +/-34 | 258 | — | — |
| 374 | zephyr-7b-alpha MIT | 1167.00 | +/-39 | 201 | — | — |
| 375 | Gemma 7B - ItGoogle Research | 1167.00 | +/-17 | 1,381 | Google Research | Gemma license |
| 376 | vicuna-13b Llama 2 Community | 1164.00 | +/-14 | 2,389 | — | — |
| 377 |  Baichuan2-13B-Chat百川智能 | 1163.00 | +/-13 | 2,626 | 百川智能 | Llama 2 Community |
| 378 | smollm2-1.7b-instruct Apache 2.0 | 1160.00 | +/-33 | 352 | — | — |
| 379 | CodeLLaMA-34BFacebook AI研究实验室 | 1160.00 | +/-20 | 853 | Facebook AI研究实验室 | Llama 2 Community |
| 380 | Phi-3-mini 3.8BMicrosoft Azure | 1155.00 | +/-13 | 3,886 | Microsoft Azure | MIT |
| 381 | PaLM 2Google Research | 1154.00 | +/-21 | 917 | Google Research | Proprietary |
| 382 | zephyr-7b-beta MIT | 1153.00 | +/-18 | 1,250 | — | — |
| 383 | wizardlm-13bMicrosoft | 1152.00 | +/-22 | 735 | Microsoft | Llama 2 Community |
| 384 | llama-3.2-1b-instructMeta | 1149.00 | +/-16 | 1,346 | Meta | Llama 3.2 |
| 385 | llama2-70b-steerlm-chatNvidia | 1146.00 | +/-28 | 467 | Nvidia | Llama 2 Community |
| 386 | Mistral 7B InstructMistralAI | 1145.00 | +/-20 | 1,032 | MistralAI | Apache 2.0 |
| 387 | Gemma 2B - ItGoogle Research | 1136.00 | +/-22 | 742 | Google Research | Gemma license |
| 388 | vicuna-7b Llama 2 Community | 1132.00 | +/-23 | 726 | — | — |
| 389 | Qwen1.5-4B-Chat阿里巴巴 | 1132.00 | +/-17 | 1,283 | 阿里巴巴 | Qianwen LICENSE |
| 390 | stripedhyena-nous-7b Apache 2.0 | 1128.00 | +/-22 | 704 | — | — |
| 391 | guanaco-33b Non-commercial | 1114.00 | +/-35 | 263 | — | — |
| 392 | olmo-7b-instructAi2 | 1107.00 | +/-22 | 772 | Ai2 | Apache-2.0 |
| 393 | Baichuan2-7B-Chat百川智能 | 1103.00 | +/-14 | 1,956 | 百川智能 | Llama 2 Community |
| 394 | ChatGLM3-6B智谱AI | 1091.00 | +/-26 | 535 | 智谱AI | — |
| 395 | mpt-7b-chat CC-BY-NC-SA-4.0 | 1067.00 | +/-31 | 397 | — | — |
| 396 |  Koala达摩院 | 1066.00 | +/-24 | 747 | 达摩院 | — |
| 397 | RWKV-4-Raven-14B Apache 2.0 | 1060.00 | +/-27 | 505 | — | — |
| 398 | oasst-pythia-12b Apache 2.0 | 1051.00 | +/-25 | 714 | — | — |
| 399 | ChatGLM-6B智谱AI | 1036.00 | +/-27 | 551 | 智谱AI | — |
| 400 | ChatGLM2-6B智谱AI | 1031.00 | +/-34 | 293 | 智谱AI | — |
| 401 | stablelm-tuned-alpha-7b CC-BY-NC-SA-4.0 | 1005.00 | +/-32 | 363 | — | — |
| 402 | alpaca-13b Non-commercial | 1000.00 | +/-27 | 626 | — | — |
| 403 | dolly-v2-12b MIT | 963.00 | +/-33 | 396 | — | — |
| 404 | fastchat-t5-3b Apache 2.0 | 908.00 | +/-30 | 428 | — | — |
| 405 | LLaMA 13BFacebook AI研究实验室 | 883.00 | +/-38 | 304 | Facebook AI研究实验室 | Non-commercial |
数据仅供参考，以官方来源为准。模型名称旁的链接可跳转到 DataLearner 模型详情页。
常见问题 (FAQ)
01
什么是 LMArena Coding Arena？
LMArena Coding Arena 是 LMArena 旗下专注于代码能力的匿名评测平台。用户提交真实编程任务（如调试、代码生成、算法实现），系统将不同模型的输出并排展示（隐藏模型名称），由用户投票选出更好的答案，最终通过 Elo 算法汇总形成动态排行榜。
02
Coding Arena 与 SWE-bench、HumanEval 等静态基准有什么区别？
SWE-bench、HumanEval、MBPP 等静态基准使用固定测试集和自动化评分，可重现性强但容易被针对性优化（"刷榜"）。Coding Arena 来自真实用户的开放式需求，测试内容不固定，更能反映模型在实际编程场景中的表现，两者互为补充。
03
国产大模型在代码能力方面表现如何？
DeepSeek、Qwen 等国产模型在 Coding Arena 表现亮眼，已跻身全球前列。DeepSeek 以 MIT 协议开源，Qwen 系列支持中文编程场景，是开发者选择开源代码模型的重要参考。
04
如何用 AI 辅助日常编程工作？
常见场景包括：代码补全与生成、调试、代码审查、单元测试生成，以及跨语言翻译。