GDPval-AA v2
高经济价值、可交付成果导向的真实知识工作任务。
9 项评测
拆解模型在知识、推理、编程、工具使用和长上下文上的差异。
INTELLIGENCE INDEX V4.1
高经济价值、可交付成果导向的真实知识工作任务。
多轮客户交互、知识库检索与工具调用。
在真实终端环境中完成构建、配置、调试和运维任务。
科研场景中的代码生成、执行和数值计算。
跨长上下文定位、保持和综合关键信息。
同时衡量知识准确性、拒答判断和幻觉风险。
覆盖多学科专家级知识与复杂推理。
研究生水平、难以通过简单检索回答的科学问题。
要求严谨分析与多步骤推理的科学评测。
完整数据
直接展示来源页面公开数据中的全部模型,不截断为精选榜单。
智能指数带 * 为原站估算值;其余百分比列已统一换算为百分数。GDPval-AA v2 与 AA-Omniscience 保留原始分值口径。
| # | 模型 | 机构 | 智能指数 | GDPval-AA v2 | τ³-Banking | Terminal-Bench v2.1 | SciCode | AA-LCR | AA-Omniscience | HLE | GPQA Diamond | CritPt | 权重 | 推理 | 发布日期 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5 | Anthropic | 59.86 | 1,759.6 | 26.8% | 84.6% | 60.2% | 70% | 40.15 | 53.3% | 92.6% | 28.6% | 闭源 | 是 | 2026-06-09 |
| 2 | GPT-5.6 Sol (max) | OpenAI | 58.89 | 1,742.61 | 33% | 88% | 56.1% | 73.7% | 21.7 | 47.2% | 94.1% | 32.3% | 闭源 | 是 | 2026-07-09 |
| 3 | GPT-5.6 Sol (xhigh) | OpenAI | 57.65 | 1,702.21 | 32.6% | 89.5% | 56% | 71% | 20.55 | 44.7% | 93.1% | 28.6% | 闭源 | 是 | 2026-07-09 |
| 4 | Kimi K3 | Kimi | 57.11 | 1,679.21 | 33.4% | 85% | 58.7% | 74.7% | 18.42 | 44.3% | 93.5% | 23.4% | 闭源 | 是 | 2026-07-16 |
| 5 | GPT-5.6 Sol (high) | OpenAI | 55.87 | 1,630.29 | 30.6% | 87.3% | 56.9% | 68.3% | 19.75 | 44.1% | 92.8% | 25.7% | 闭源 | 是 | 2026-07-09 |
| 6 | Claude Opus 4.8 (max) | Anthropic | 55.69 | 1,600.12 | 27.6% | 84.6% | 53.5% | 67.7% | 27.43 | 45.7% | 92% | 20.9% | 闭源 | 是 | 2026-05-28 |
| 7 | GPT-5.6 Terra (max) | OpenAI | 54.95 | 1,581.23 | 31.8% | 88% | 53.9% | 74% | -0.22 | 41.8% | 92.5% | 30% | 闭源 | 是 | 2026-07-09 |
| 8 | GPT-5.5 (xhigh)历史 | OpenAI | 54.84 | 1,493.18 | 31.3% | 84.3% | 56.1% | 74.3% | 20.07 | 44.3% | 93.5% | 27.1% | 闭源 | 是 | 2026-04-23 |
| 9 | Grok 4.5 (high) | SpaceXAI | 53.83 | 1,535 | 32.6% | 81.6% | 54.1% | 67.7% | 26.38 | 40.3% | 93.1% | 15.4% | 闭源 | 是 | 2026-07-08 |
| 10 | GPT-5.6 Sol (medium) | OpenAI | 53.59 | 1,561.65 | 26.5% | 86.1% | 56.5% | 68.7% | 18.95 | 39.7% | 92.6% | 22.9% | 闭源 | 是 | 2026-07-09 |
| 11 | Claude Opus 4.7 (max)历史 | Anthropic | 53.53 | 1,499.64 | 28.9% | 83.1% | 54.5% | 70.3% | 26.17 | 39.6% | 91.4% | 12% | 闭源 | 是 | 2026-04-16 |
| 12 | Claude Sonnet 5 (max) | Anthropic | 53.35 | 1,607 | 28.2% | 80.5% | 53.6% | 70.7% | 15.32 | 39.6% | 91.1% | 16.9% | 闭源 | 是 | 2026-06-30 |
| 13 | GPT-5.5 (high)历史 | OpenAI | 53.13 | 1,471.02 | 29.5% | 79.4% | 55.9% | 73.3% | 18.37 | 43% | 93.2% | 25.4% | 闭源 | 是 | 2026-04-23 |
| 14 | GPT-5.6 Terra (xhigh) | OpenAI | 51.6 | 1,571.97 | 24.3% | 80.1% | 51.6% | 71.3% | -3.3 | 40% | 90.8% | 27.1% | 闭源 | 是 | 2026-07-09 |
| 15 | GPT-5.4 (xhigh)历史 | OpenAI | 51.4 | 1,394.68 | 30.3% | 78.3% | 56.6% | 74% | 5.65 | 41.6% | 92% | 23.4% | 闭源 | 是 | 2026-03-05 |
| 16 | GPT-5.6 Luna (max) | OpenAI | 51.24 | 1,584 | 27.2% | 80.9% | 52.5% | 74% | -11.23 | 37.2% | 91.1% | 20.6% | 闭源 | 是 | 2026-07-09 |
| 17 | GLM-5.2 (max) | Z AI | 51.09 | 1,513.89 | 26.8% | 77.9% | 50.5% | 71.3% | 3.97 | 40.1% | 89.5% | 20.9% | 开放 | 是 | 2026-06-16 |
| 18 | Muse Spark 1.1 (xhigh) | Meta | 50.62 | 1,374.33 | 25.2% | 77.9% | 58.2% | 63.3% | 18 | 45.1% | 89.8% | 15.1% | 闭源 | 是 | 2026-07-09 |
| 19 | GPT-5.5 (medium)历史 | OpenAI | 50.41 | 1,374.63 | 25.8% | 80.5% | 53.5% | 72.3% | 17.28 | 40.6% | 92.6% | 18.6% | 闭源 | 是 | 2026-04-23 |
| 20 | Gemini 3.5 Flash | 50.2 | 1,349 | 25.4% | 78.7% | 53.1% | 69.3% | 22.68 | 41% | 92.2% | 13.1% | 闭源 | 是 | 2026-05-19 | |
| 21 | GPT-5.6 Sol (low) | OpenAI | 49.44 | 1,444.68 | 24.4% | 76.8% | 55.4% | 67.7% | 18.4 | 36.6% | 89.8% | 14.9% | 闭源 | 是 | 2026-07-09 |
| 22 | GPT-5.6 Luna (xhigh) | OpenAI | 49.07 | 1,539.11 | 24.3% | 77.9% | 50% | 69.7% | -11.6 | 35.6% | 89.5% | 20.6% | 闭源 | 是 | 2026-07-09 |
| 23 | GPT-5.6 Terra (high) | OpenAI | 48.95 | 1,512.84 | 22.3% | 75.7% | 50.1% | 72.3% | -4.07 | 36.7% | 89.6% | 22.9% | 闭源 | 是 | 2026-07-09 |
| 24 | Claude Sonnet 4.6 (max)历史 | Anthropic | 47.21 | 1,377.16 | 30.5% | 71.2% | 46.8% | 70.7% | 12.37 | 30% | 87.5% | 3.1% | 闭源 | 是 | 2026-02-17 |
| 25 | Gemini 3.1 Pro Preview | 46.46 | 965 | 16.5% | 73.8% | 58.9% | 72.7% | 32.93 | 44.7% | 94.1% | 17.7% | 闭源 | 是 | 2026-02-19 | |
| 26 | GPT-5.6 Luna (high) | OpenAI | 46.06 | 1,472.36 | 22.3% | 69.7% | 50.7% | 69% | -12.45 | 31.6% | 89.2% | 16.6% | 闭源 | 是 | 2026-07-09 |
| 27 | Qwen3.7 Max | Alibaba | 45.99 | 1,273.15 | 10.9% | 74.5% | 48.8% | 69% | 14.08 | 38.1% | 92.3% | 13.4% | 闭源 | 是 | 2026-05-19 |
| 28 | GPT-5.6 Terra (medium) | OpenAI | 45.57 | 1,404.45 | 19.4% | 72.3% | 49.7% | 68% | -5.17 | 31.6% | 87.2% | 17.4% | 闭源 | 是 | 2026-07-09 |
| 29 | Gemini 3.5 Flash (medium) | 45.38* | — | — | — | 53% | 71% | 21.65 | 39.9% | 92.1% | 10.9% | 闭源 | 是 | 2026-05-19 | |
| 30 | MiniMax-M3 | MiniMax | 44.44 | 1,394.83 | 13% | 65.2% | 45.4% | 74% | 1.37 | 37.1% | 92.9% | 3.7% | 开放 | 是 | 2026-06-01 |
| 31 | GPT-5.3 Codex (xhigh) | OpenAI | 44.27* | — | — | — | 53.2% | 74% | 9.88 | 39.9% | 91.5% | 16.9% | 闭源 | 是 | 2026-02-05 |
| 32 | DeepSeek V4 Pro (max) | DeepSeek | 44.27 | 1,307.21 | 25.8% | 64% | 50% | 66.3% | -10.02 | 35.9% | 88.8% | 12.9% | 开放 | 是 | 2026-04-24 |
| 33 | Kimi K2.6历史 | Kimi | 44.22 | 1,189.4 | 20.6% | 65.9% | 53.5% | 69.7% | 6.42 | 35.9% | 91.1% | 8% | 开放 | 是 | 2026-04-20 |
| 34 | Motif 3 (Beta) | Motif Technologies | 44.08 | 1,259.04 | 28.9% | 70.8% | 44.3% | 61.3% | -17.28 | 38.2% | 86.9% | 6% | 闭源 | 是 | — |
| 35 | Claude Opus 4.6 (max)历史 | Anthropic | 43.71* | — | — | — | 51.9% | 70.7% | 13.5 | 36.7% | 89.6% | 12.6% | 闭源 | 是 | 2026-02-05 |
| 36 | GPT-5.5 (low)历史 | OpenAI | 43.46 | 1,190.84 | 21.2% | 65.5% | 51.6% | 72% | 14.62 | 31% | 91% | 8% | 闭源 | 是 | 2026-04-23 |
| 37 | DeepSeek V4 Pro (high) | DeepSeek | 43.11 | 1,298.76 | 23.7% | 64.8% | 46.4% | 65% | -9.7 | 33.5% | 90.5% | 10% | 开放 | 是 | 2026-04-24 |
| 38 | Muse Spark | Meta | 43.06 | 1,143.69 | 19.6% | 62.2% | 51.5% | 69.7% | 4.08 | 39.9% | 88.4% | 11.3% | 闭源 | 是 | 2026-04-08 |
| 39 | Claude Opus 4.7 (Non-reasoning, high) | Anthropic | 42.68* | — | — | — | 50.1% | 67% | 14.22 | 31.2% | 88.5% | 5.1% | 闭源 | 否 | 2026-04-16 |
| 40 | MiMo-V2.5-Pro | Xiaomi | 42.24 | 1,265.4 | 8.7% | 65.2% | 50.2% | 73.3% | 3.6 | 33.8% | 86.6% | 4% | 开放 | 是 | 2026-04-22 |
| 41 | GPT-5.2 (xhigh)历史 | OpenAI | 42.18* | — | — | — | 52.1% | 72.7% | -1 | 35.4% | 90.3% | 11.6% | 闭源 | 是 | 2025-12-11 |
| 42 | Kimi K2.7 Code | Kimi | 41.95 | 1,186.79 | 18.1% | 67.4% | 47.5% | 66.3% | -10.7 | 32.8% | 89.6% | 10% | 开放 | 是 | 2026-06-12 |
| 43 | Claude Sonnet 5 (Non-reasoning) | Anthropic | 41.73 | 1,371.85 | 14% | 75.3% | 48.6% | 58.7% | -1.18 | 17.8% | 80% | 1.1% | 闭源 | 否 | 2026-06-30 |
| 44 | Hy3 | Tencent | 41.23 | 1,213.87 | 20.8% | 64.4% | 47.6% | 66.7% | -18.5 | 31.6% | 89.7% | 4.9% | 开放 | 是 | 2026-07-06 |
| 45 | GPT-5.6 Sol (Non-reasoning) | OpenAI | 41.2 | 1,383.93 | 16.1% | 74.2% | 47.1% | 55.3% | 0.85 | 15.8% | 79% | 5.1% | 闭源 | 否 | 2026-07-09 |
| 46 | Nex-N2-Pro | Nex AGI | 40.96 | 1,256.87 | 16.9% | 67.8% | 41.8% | 67.7% | -27.62 | 32.4% | 89.2% | 8.6% | 开放 | 是 | 2026-06-02 |
| 47 | Claude Opus 4.5历史 | Anthropic | 40.77* | — | — | — | 49.5% | 74% | 13.27 | 28.4% | 86.6% | 4.6% | 闭源 | 是 | 2025-11-24 |
| 48 | Inkling | Thinking Machines | 40.74 | 1,238.85 | 23.7% | 55.1% | 46.1% | 63.3% | 2.05 | 29.7% | 87.2% | 5.4% | 开放 | 是 | 2026-07-15 |
| 49 | GPT-5.6 Terra (low) | OpenAI | 40.47 | 1,248.54 | 16.1% | 62.5% | 49.2% | 64.3% | -7.12 | 27.4% | 84.3% | 9.4% | 闭源 | 是 | 2026-07-09 |
| 50 | MiMo-V2-Pro历史 | Xiaomi | 40.29* | — | — | — | 42.5% | 60.7% | 4.92 | 28.3% | 87% | 0.3% | 闭源 | 是 | 2026-03-18 |
| 51 | DeepSeek V4 Flash (max) | DeepSeek | 40.28 | 1,188.54 | 22.9% | 61.8% | 44.9% | 63% | -22.9 | 32.1% | 89.4% | 7.1% | 开放 | 是 | 2026-04-24 |
| 52 | GLM-5.1历史 | Z AI | 40.16 | 1,256.68 | 11.5% | 61.8% | 43.8% | 62.3% | 1.93 | 28% | 86.8% | 4.6% | 开放 | 是 | 2026-04-07 |
| 53 | GPT-5.2 Codex (xhigh)历史 | OpenAI | 40.14* | — | — | — | 54.6% | 75.7% | -2.48 | 33.5% | 89.9% | 8.7% | 闭源 | 是 | 2025-12-11 |
| 54 | Qwen3.6 Max Preview历史 | Alibaba | 40* | — | — | — | 46.9% | 69.7% | 10.2 | 28.9% | 88.8% | 3.7% | 闭源 | 是 | 2026-04-20 |
| 55 | GPT-5.4 mini (xhigh)历史 | OpenAI | 39.98 | 1,171 | 21.4% | 59.2% | 49.9% | 69.3% | -18.68 | 26.6% | 87.5% | 10% | 闭源 | 是 | 2026-03-17 |
| 56 | Grok Build 0.1 0616历史 | SpaceXAI | 39.8 | 1,211.39 | 10.9% | 52.1% | 50.2% | 64.7% | 7.43 | 36% | 89.5% | 9.1% | 闭源 | 是 | 2026-06-16 |
| 57 | Qwen3.6 Plus | Alibaba | 39.56 | 1,135.39 | 16.5% | 61.4% | 40.7% | 69.7% | 2.65 | 25.7% | 88.2% | 2.9% | 闭源 | 是 | 2026-04-02 |
| 58 | Gemini 3 Pro Preview (high)历史 | 39.55* | — | — | — | 56.1% | 70.7% | 15.8 | 37.2% | 90.8% | 9.1% | 闭源 | 是 | 2025-11-18 | |
| 59 | GLM-5历史 | Z AI | 39.5* | — | — | — | 46.2% | 63.3% | 2 | 27.2% | 82% | 2% | 开放 | 是 | 2026-02-11 |
| 60 | GPT-5.4 (low)历史 | OpenAI | 39.14* | — | — | — | 50.3% | 67.3% | 4.72 | 28.9% | 87.1% | 7.4% | 闭源 | 是 | 2026-03-05 |
| 61 | Qwen3.7 Plus | Alibaba | 38.98 | 936.26 | 17.9% | 61% | 45.5% | 65% | 2.37 | 33.4% | 90% | 9.1% | 闭源 | 是 | 2026-06-01 |
| 62 | JT-4.1 Flash 236B A21B | China Mobile | 38.83 | 1,264.54 | 28% | 59.6% | 38.2% | 58.7% | -10.12 | 16.1% | 84.5% | 0% | 闭源 | 否 | 2026-07-09 |
| 63 | GPT-5.4 nano (xhigh)历史 | OpenAI | 38.24 | 1,099.89 | 21% | 60.7% | 46.9% | 66% | -29.55 | 26.5% | 81.7% | 9.3% | 闭源 | 是 | 2026-03-17 |
| 64 | MiniMax-M2.7历史 | MiniMax | 38.13 | 1,157.91 | 8.9% | 55.4% | 47% | 68.7% | 0.68 | 28.1% | 87.4% | 0.6% | 开放 | 是 | 2026-03-18 |
| 65 | GLM-5-Turbo历史 | Z AI | 38.06* | — | — | — | 43.6% | 60.7% | -15.08 | 25.4% | 84.7% | 0.3% | 闭源 | 是 | 2026-03-15 |
| 66 | GPT-5.6 Luna (medium) | OpenAI | 38.05 | 1,272.37 | 15.3% | 53.2% | 45.8% | 66% | -13.82 | 24.5% | 85.9% | 4.9% | 闭源 | 是 | 2026-07-09 |
| 67 | GPT-5.2 (medium)历史 | OpenAI | 37.95* | — | — | — | 46.2% | 63.3% | 0.02 | 24.9% | 86.4% | 7.9% | 闭源 | 是 | 2025-12-11 |
| 68 | Claude Opus 4.6 (high)历史 | Anthropic | 37.79* | — | — | — | 45.7% | 58.3% | 3.47 | 18.6% | 84% | 2.8% | 闭源 | 否 | 2026-02-05 |
| 69 | Gemini 3 Flash历史 | 37.76* | — | 17.5% | — | 50.6% | 66.3% | 11.57 | 34.7% | 89.8% | 8.6% | 闭源 | 是 | 2025-12-17 | |
| 70 | Nemotron 3 Ultra | NVIDIA | 37.76 | 1,163.8 | 13.8% | 53.9% | 39.9% | 67% | -0.83 | 26.6% | 86.7% | 3.1% | 开放 | 是 | 2026-06-04 |
| 71 | Grok 4.3 (high)历史 | SpaceXAI | 37.58 | 1,084.9 | 12.2% | 39.7% | 47.3% | 64.3% | 18.32 | 35% | 90.1% | 8% | 闭源 | 是 | 2026-04-30 |
| 72 | DeepSeek V4 Flash (high) | DeepSeek | 37.36* | — | — | — | 42% | 62.7% | -22.28 | 27.8% | 86.7% | 3.4% | 开放 | 是 | 2026-04-24 |
| 73 | MiMo-V2.5 | Xiaomi | 37.2 | 1,145.01 | 6.6% | 63.7% | 43.1% | 62.7% | -9.33 | 25.2% | 84.9% | 3.7% | 开放 | 是 | 2026-04-22 |
| 74 | Qwen3.6 27B | Alibaba | 37.05 | 1,139.72 | 15.3% | 60.7% | 39.8% | 68.7% | -19.78 | 21.6% | 84.2% | 1.1% | 开放 | 是 | 2026-04-22 |
| 75 | Grok 4.20 0309 v2历史 | SpaceXAI | 37* | — | — | — | 45.6% | 58% | 15.35 | 32.2% | 91.1% | 6.6% | 闭源 | 是 | 2026-04-07 |
| 76 | GPT-5.1 (high)历史 | OpenAI | 36.87 | 987.45 | 14% | 52.4% | 43.3% | 75% | 5.55 | 26.5% | 87.3% | 4.9% | 闭源 | 是 | 2025-11-13 |
| 77 | Grok 4.20 0309历史 | SpaceXAI | 36.5* | — | — | — | 44.7% | 59% | 13.37 | 30% | 88.5% | 6% | 闭源 | 是 | 2026-03-10 |
| 78 | Claude 4.5 Sonnet历史 | Anthropic | 36.42 | 1,052.46 | 19% | 55.8% | 44.7% | 65.7% | 0.33 | 17.3% | 83.4% | 1.1% | 闭源 | 是 | 2025-09-29 |
| 79 | MiMo-V2-Omni-0327 | Xiaomi | 36.39* | — | — | — | 39.5% | 63.7% | -14.02 | 20.4% | 85.5% | 0.9% | 闭源 | 是 | 2026-03-27 |
| 80 | GPT-5 Codex (high)历史 | OpenAI | 36.12* | — | — | — | 40.9% | 69% | -6.77 | 25.6% | 83.7% | 5.1% | 闭源 | 是 | 2025-09-23 |
| 81 | Grok 4.3 (medium) | SpaceXAI | 36* | — | — | — | 44.6% | 65% | 16.98 | 28.1% | 89% | 4.9% | 闭源 | 是 | 2026-04-30 |
| 82 | Claude Sonnet 4.6 (Non-reasoning)历史 | Anthropic | 35.89* | — | — | — | 46.9% | 57.7% | -2.93 | 13.2% | 79.9% | 0.9% | 闭源 | 否 | 2026-02-17 |
| 83 | Grok 4.3 (low) | SpaceXAI | 35.44* | — | — | — | 41.9% | 64% | 13.67 | 17.3% | 84.3% | 0.6% | 闭源 | 是 | 2026-04-30 |
| 84 | GPT-5.5 (Non-reasoning)历史 | OpenAI | 35.38 | 1,118.97 | 13.8% | 61% | 47.3% | 56.3% | -5.05 | 12.6% | 76.8% | 1.4% | 闭源 | 否 | 2026-04-23 |
| 85 | Kimi K2.5历史 | Kimi | 35.37 | 1,008.53 | 14.2% | 45.7% | 49% | 65.3% | -8.12 | 29.4% | 87.9% | 3.1% | 开放 | 是 | 2026-01-27 |
| 86 | GLM-5.1历史 | Z AI | 35.37* | — | — | — | 36.1% | 44.3% | -21.4 | 25.6% | 83.9% | 0% | 开放 | 否 | 2026-04-07 |
| 87 | MiMo-V2-Omni | Xiaomi | 34.99* | — | — | — | 36.7% | 66.7% | -17.43 | 19.9% | 82.8% | 1.1% | 闭源 | 是 | 2026-03-19 |
| 88 | Gemini 3.5 Flash (minimal) | 34.91* | — | — | — | 48.8% | 53.3% | 1.3 | 23.1% | 82.8% | 1.4% | 闭源 | 否 | 2026-05-19 | |
| 89 | GPT-5.1 Codex (high)历史 | OpenAI | 34.73* | — | — | — | 40.2% | 67.3% | -6.02 | 23.4% | 86% | 5.7% | 闭源 | 是 | 2025-11-13 |
| 90 | Claude Opus 4.5历史 | Anthropic | 34.71* | — | — | — | 47% | 65.3% | -3.95 | 12.9% | 81% | 0.3% | 闭源 | 否 | 2025-11-24 |
| 91 | GPT-5 (high)历史 | OpenAI | 34.71 | 1,074.8 | 19.6% | 35.2% | 42.9% | 75.6% | -8.08 | 26.5% | 85.4% | 5.7% | 闭源 | 是 | 2025-08-07 |
| 92 | Kimi K2.6 | Kimi | 34.58* | — | — | — | 39.5% | 57.7% | -9.97 | 18.2% | 78.8% | 1.4% | 开放 | 否 | 2026-04-20 |
| 93 | GLM 5V Turbo历史 | Z AI | 34.49* | — | — | — | 43.5% | 61% | -18.98 | 15.8% | 80.9% | 0.6% | 闭源 | 是 | 2026-04-01 |
| 94 | Claude Sonnet 4.6 (Non-reasoning, Low Effort) | Anthropic | 34.26* | — | — | — | 44.1% | 58.7% | -1.9 | 10.8% | 79.7% | 0.9% | 闭源 | 否 | 2026-02-17 |
| 95 | GLM-5.2 | Z AI | 34.05 | 1,387.57 | 15.7% | 51.7% | 36.1% | 37.3% | -6.4 | 8.4% | 68.6% | 3.1% | 开放 | 否 | 2026-06-16 |
| 96 | GPT-5.6 Terra (Non-reasoning) | OpenAI | 33.97 | 1,238.72 | 13.4% | 56.2% | 44.6% | 50% | -23.47 | 11% | 74.6% | 2% | 闭源 | 否 | 2026-07-09 |
| 97 | Qwen3.5 27B历史 | Alibaba | 33.78* | — | — | — | 39.5% | 67.3% | -42.02 | 22.2% | 85.8% | 0.9% | 开放 | 是 | 2026-02-24 |
| 98 | GPT-5 (medium)历史 | OpenAI | 33.74* | — | — | — | 41.1% | 72.8% | -10.08 | 23.5% | 84.2% | 0% | 闭源 | 是 | 2025-08-07 |
| 99 | GLM-4.7历史 | Z AI | 33.7 | 1,165.3 | 9.7% | 45.3% | 45.1% | 64% | -34.6 | 25.1% | 85.9% | 1.7% | 开放 | 是 | 2025-12-22 |
| 100 | KAT-Coder-Pro V2 | KwaiKAT | 33.69 | 908.18 | 5.8% | 70% | 38.3% | 66% | -21.6 | 16% | 85.5% | 0% | 闭源 | 否 | 2026-03-27 |
| 101 | Qwen3.5 397B A17B | Alibaba | 33.68 | 961.91 | 13.4% | 51.3% | 42% | 65.7% | -29.78 | 27.3% | 89.3% | 1.7% | 开放 | 是 | 2026-02-16 |
| 102 | MiniMax-M2.5历史 | MiniMax | 33.65* | — | — | — | 42.6% | 66% | -39.7 | 19.1% | 84.8% | 1.1% | 开放 | 是 | 2026-02-12 |
| 103 | Hy3-preview | Tencent | 33.58* | — | — | — | 41.2% | 54.7% | -34.63 | 25.5% | 86.7% | 4.6% | 开放 | 是 | 2026-04-23 |
| 104 | GPT-5.5 Instant (May 2026)历史 | OpenAI | 33.52* | — | — | — | 50.3% | 55.7% | 10.33 | 20.3% | 84.6% | 2.6% | 闭源 | 是 | 2026-05-05 |
| 105 | LongCat 2.0 | LongCat | 33.48 | 1,021.84 | 12.8% | 50.2% | 35.4% | 58% | -22.63 | 32.1% | 78% | 2.6% | 开放 | 是 | 2026-06-29 |
| 106 | Grok 4历史 | SpaceXAI | 33.28* | — | — | — | 45.7% | 68% | 3.75 | 23.9% | 87.7% | 2% | 闭源 | 是 | 2025-07-10 |
| 107 | GPT-5.6 Luna (low) | OpenAI | 33.26 | 1,150.08 | 12% | 43.4% | 45.6% | 59.3% | -15.2 | 18.8% | 83.5% | 2.6% | 闭源 | 是 | 2026-07-09 |
| 108 | MiMo-V2-Flash (Feb 2026) | Xiaomi | 33.22* | — | — | — | 38.3% | 64.3% | -18.47 | 20% | 83.5% | 2.9% | 开放 | 是 | 2025-12-16 |
| 109 | Gemini 3 Pro Preview (low)历史 | 33.07* | — | — | — | 49.9% | 67.3% | 2.15 | 27.6% | 88.7% | 0% | 闭源 | 是 | 2025-11-18 | |
| 110 | Kimi K2 Thinking历史 | Kimi | 32.7* | — | — | — | 42.4% | 66.3% | -20.48 | 22.3% | 83.8% | 2.6% | 开放 | 是 | 2025-11-06 |
| 111 | GLM-5历史 | Z AI | 32.41* | — | — | — | 38.3% | 37% | -11.23 | 7.2% | 66.6% | 0% | 开放 | 否 | 2026-02-11 |
| 112 | Qwen3.5 122B A10B | Alibaba | 32.28 | 978.02 | 13.6% | 47.6% | 42% | 66.7% | -39.58 | 23.4% | 85.7% | 0.6% | 开放 | 是 | 2026-02-24 |
| 113 | DeepSeek V3.2历史 | DeepSeek | 32.04 | 865.48 | 18.8% | 46.8% | 38.9% | 65% | -20.88 | 22.2% | 84% | 2.9% | 开放 | 是 | 2025-12-01 |
| 114 | Qwen3.5 397B A17B | Alibaba | 31.98* | — | — | — | 41.1% | 58% | -36.08 | 18.8% | 86.1% | 0.9% | 开放 | 否 | 2026-02-16 |
| 115 | Qwen3 Max Thinking历史 | Alibaba | 31.75* | — | — | — | 43.1% | 66% | -34.42 | 26.2% | 86.1% | 1.7% | 闭源 | 是 | 2026-01-26 |
| 116 | Qwen3.6 35B A3B | Alibaba | 31.65 | 1,048.67 | 8.7% | 44.9% | 35.8% | 63.7% | -21.43 | 20.2% | 84.1% | 0.3% | 开放 | 是 | 2026-04-16 |
| 117 | MiniMax-M2.1历史 | MiniMax | 31.36* | — | — | — | 40.7% | 59% | -32.83 | 22.2% | 83% | 0.3% | 开放 | 是 | 2025-12-23 |
| 118 | DeepSeek V4 Pro | DeepSeek | 31.22* | — | — | — | 42.4% | 45% | -30.08 | 7.7% | 71.7% | 0.9% | 开放 | 否 | 2026-04-24 |
| 119 | MiMo-V2-Flash历史 | Xiaomi | 31.2* | — | — | — | 39.4% | 63% | -43.45 | 21.1% | 84.6% | 4.3% | 开放 | 是 | 2025-12-16 |
| 120 | GPT-5 (low)历史 | OpenAI | 31.15* | — | — | — | 39.1% | 58.7% | -10.02 | 18.4% | 80.8% | 1.1% | 闭源 | 是 | 2025-08-07 |
| 121 | GPT-5 mini (medium)历史 | OpenAI | 30.92* | — | — | — | 41% | 66% | -10.78 | 14.6% | 80.3% | 1.4% | 闭源 | 是 | 2025-08-07 |
| 122 | Qwen3.5 Omni Plus | Alibaba | 30.64* | — | — | — | 40.5% | 52.7% | -12.3 | 13.9% | 82.6% | 0.6% | 闭源 | 否 | 2026-03-30 |
| 123 | GPT-5.1 Codex mini (high)历史 | OpenAI | 30.63* | — | — | — | 42.6% | 62.7% | -16.52 | 16.9% | 81.3% | 0% | 闭源 | 是 | 2025-11-13 |
| 124 | Grok 4.1 Fast历史 | SpaceXAI | 30.62* | — | — | — | 44.2% | 68% | -28.73 | 17.6% | 85.3% | 2.9% | 闭源 | 是 | 2025-11-19 |
| 125 | Ring-2.6-1T | InclusionAI | 30.57 | 918.78 | 14.2% | 43.1% | 42.4% | 64.3% | -37.75 | 18.3% | 85.7% | 3.7% | 开放 | 是 | 2026-05-08 |
| 126 | Qwen3.6 27B | Alibaba | 30.46 | 1,112.72 | 8.2% | 51.3% | 37.3% | 55% | -52.78 | 13.6% | 82.9% | 0.9% | 开放 | 否 | 2026-04-22 |
| 127 | DeepSeek V3.1 Terminus历史 | DeepSeek | 30.44 | 889.28 | 15.9% | 44.9% | 40.6% | 65% | -24.32 | 15.2% | 79.2% | 1.7% | 开放 | 是 | 2025-09-22 |
| 128 | o3 | OpenAI | 30.4* | — | — | — | 41% | 69.3% | -15.27 | 20% | 82.7% | 1.1% | 闭源 | 是 | 2025-04-16 |
| 129 | Step 3.7 Flash | StepFun | 30.27 | 1,017.04 | 11.3% | 39.3% | 40% | 63.7% | -37.55 | 19.9% | 80.9% | 2.3% | 开放 | 是 | 2026-05-29 |
| 130 | GPT-5.4 nano历史 | OpenAI | 30.16* | — | — | — | 38.4% | 57.3% | -18.82 | 14.7% | 76.1% | 5.1% | 闭源 | 是 | 2026-03-17 |
| 131 | Mistral Medium 3.5 | Mistral | 29.95 | 928.95 | 14.4% | 50.6% | 39.6% | 61% | -36.32 | 12.8% | 74.8% | 0% | 开放 | 是 | 2026-04-29 |
| 132 | GPT-5.4 mini (medium)历史 | OpenAI | 29.81* | — | — | — | 44.2% | 61.3% | -20.12 | 17.1% | 82.3% | 2.9% | 闭源 | 是 | 2026-03-17 |
| 133 | Claude 4.5 Haiku | Anthropic | 29.58 | 907 | 9.1% | 44.2% | 43.3% | 70.3% | -4.22 | 9.7% | 67.2% | 0% | 闭源 | 是 | 2025-10-15 |
| 134 | Kimi K2.5历史 | Kimi | 29.4* | — | — | — | 39.6% | 59% | -14 | 12.3% | 78.9% | 0.6% | 开放 | 否 | 2026-01-27 |
| 135 | Gemma 4 31B | 29.35 | 804.03 | 15.1% | 43.4% | 43.4% | 62% | -45.42 | 22.7% | 85.7% | 1.4% | 开放 | 是 | 2026-04-02 | |
| 136 | Qwen3.5 27B历史 | Alibaba | 29.31* | — | — | — | 36.7% | 55.7% | -46.75 | 13.2% | 84.2% | 0.3% | 开放 | 否 | 2026-02-24 |
| 137 | Claude 4.5 Sonnet历史 | Anthropic | 29.28* | — | — | — | 42.8% | 51.3% | -9.2 | 7.1% | 72.7% | 0% | 闭源 | 否 | 2025-09-29 |
| 138 | Qwen3.5 35B A3B历史 | Alibaba | 29.26* | — | — | — | 37.7% | 62.7% | -46.38 | 19.7% | 84.5% | 0.9% | 开放 | 是 | 2026-02-24 |
| 139 | Claude 4 Sonnet历史 | Anthropic | 28.92 | 860.55 | 13.8% | 36.3% | 40% | 64.7% | 0.18 | 9.6% | 77.7% | 0.3% | 闭源 | 是 | 2025-05-22 |
| 140 | GPT-5.5 Instant (June 2026) | OpenAI | 28.86 | 718.56 | 11.5% | 34.8% | 48.6% | 64% | 3.48 | 18.6% | 82.3% | 0% | 闭源 | 是 | 2026-06-25 |
| 141 | GLM-4.6历史 | Z AI | 28.7 | 933.57 | 10.5% | 49.4% | 38.4% | 54.3% | -41.73 | 13.3% | 78% | 1.1% | 开放 | 是 | 2025-09-30 |
| 142 | DeepSeek V4 Flash | DeepSeek | 28.65* | — | — | — | 37.3% | 33.3% | -44.18 | 7% | 71.6% | 0.3% | 开放 | 否 | 2026-04-24 |
| 143 | JT-35B-Flash | China Mobile | 28.36* | — | — | — | 29.1% | 55.3% | -23.25 | 6.1% | 82.9% | 0% | 闭源 | 否 | 2026-05-14 |
| 144 | MiniMax-M2历史 | MiniMax | 28.31* | — | — | — | 36.1% | 61% | -46.93 | 12.5% | 77.7% | 0.9% | 开放 | 是 | 2025-10-26 |
| 145 | KAT-Coder-Pro V1 | KwaiKAT | 28.26* | 898.59 | — | — | 36.6% | 74% | -37.43 | 33.4% | 76.4% | 0% | 闭源 | 否 | 2025-11-11 |
| 146 | MiMo-V2.5-Pro | Xiaomi | 27.86* | — | — | — | 39.1% | 35% | -38.3 | 13.3% | 76.2% | 1.1% | 开放 | 否 | 2026-04-22 |
| 147 | GPT-5.4 (Non-reasoning)历史 | OpenAI | 27.68* | — | — | — | 47.1% | 47.3% | -15.88 | 10.6% | 74.8% | 0.6% | 闭源 | 否 | 2026-03-05 |
| 148 | Qwen3.5 122B A10B | Alibaba | 27.6 | 879.91 | 9.1% | 47.2% | 35.6% | 56% | -53.9 | 14.8% | 82.7% | 0.9% | 开放 | 否 | 2026-02-24 |
| 149 | Gemini 3 Flash历史 | 27.37* | — | — | — | 49.9% | 48% | -3.62 | 14.1% | 81.2% | 1.4% | 闭源 | 否 | 2025-12-17 | |
| 150 | Grok 4 Fast历史 | SpaceXAI | 27.37* | — | — | — | 44.2% | 64.7% | -28.4 | 17% | 84.7% | 2.9% | 闭源 | 是 | 2025-09-19 |
| 151 | Claude 3.7 Sonnet历史 | Anthropic | 27.06* | — | — | — | 40.3% | 60.7% | 0.32 | 10.3% | 77.2% | 0.9% | 闭源 | 是 | 2025-02-24 |
| 152 | GPT-5.6 Luna (Non-reasoning) | OpenAI | 26.56 | 1,068.16 | 9.1% | 39% | 39.9% | 36.3% | -25.12 | 6.7% | 64.5% | 0.3% | 闭源 | 否 | 2026-07-09 |
| 153 | GLM-4.7历史 | Z AI | 26.56* | — | — | — | 35.4% | 36.3% | -46.28 | 6.1% | 66.4% | 0% | 开放 | 否 | 2025-12-22 |
| 154 | Hy3-preview | Tencent | 26.1* | — | — | — | 39.4% | 33.7% | -36.4 | 6.3% | 73.2% | 0.3% | 开放 | 否 | 2026-04-23 |
| 155 | Ling-2.6-1T | InclusionAI | 26.05* | — | — | — | 37% | 34.7% | -51 | 8.2% | 75.2% | 0.3% | 开放 | 否 | 2026-04-23 |
| 156 | GPT-5.2历史 | OpenAI | 26.02* | — | 12% | — | 40.4% | 38% | -13.72 | 7.3% | 71.2% | 0.6% | 闭源 | 否 | 2025-12-11 |
| 157 | Step 3.5 Flash 2603 | StepFun | 26* | — | — | — | 38.5% | 54.3% | -43.67 | 22.6% | 82.6% | 2.3% | 闭源 | 是 | 2026-04-02 |
| 158 | Doubao Seed Code | ByteDance Seed | 25.98* | — | — | — | 40.7% | 65.3% | -34.43 | 13.3% | 76.4% | 0.3% | 闭源 | 是 | 2025-11-11 |
| 159 | Gemini 2.5 Pro | 25.82 | 665.39 | 9.3% | 28.5% | 42.8% | 66% | -14.3 | 21.1% | 84.4% | 2.6% | 闭源 | 是 | 2025-06-05 | |
| 160 | Gemma 4 26B A4B | 25.69 | 761.33 | 11.8% | 39% | 40% | 55.7% | -48.07 | 18.3% | 79.2% | 0% | 开放 | 是 | 2026-04-02 | |
| 161 | o4-mini (high)历史 | OpenAI | 25.55* | — | — | — | 46.5% | 55% | -35.75 | 17.5% | 78.4% | 0.6% | 闭源 | 是 | 2025-04-16 |
| 162 | Step 3.5 Flash历史 | StepFun | 25.5* | — | — | — | 40.4% | 43% | -40.9 | 19.1% | 83.1% | 2.5% | 开放 | 是 | 2026-02-02 |
| 163 | Claude 4 Sonnet历史 | Anthropic | 25.49* | — | — | — | 37.3% | 44.3% | -9.23 | 4% | 68.3% | 1.1% | 闭源 | 否 | 2025-05-22 |
| 164 | DeepSeek V3.2 Exp历史 | DeepSeek | 25.45* | — | — | — | 37.7% | 69% | -29.93 | 13.8% | 79.7% | 1.4% | 开放 | 是 | 2025-09-29 |
| 165 | NVIDIA Nemotron 3 Super | NVIDIA | 25.41 | 693.25 | 10.1% | 38.6% | 36% | 60% | -42.07 | 19.2% | 80% | 3.1% | 开放 | 是 | 2026-03-11 |
| 166 | GPT-5 mini (high)历史 | OpenAI | 25.31 | 936.71 | 14.6% | 3.7% | 39.2% | 68% | -17.18 | 19.7% | 82.8% | 0% | 闭源 | 是 | 2025-08-07 |
| 167 | Gemini 3.1 Flash-Lite | 25.04 | 642 | 8.7% | 31.1% | 41.9% | 65.3% | -15.52 | 16.2% | 82.2% | 1.1% | 闭源 | 是 | 2026-03-03 | |
| 168 | Qwen3 Max Thinking (Preview)历史 | Alibaba | 25.03* | — | — | — | 38.7% | 57.7% | -37.83 | 12% | 77.6% | 0% | 闭源 | 是 | 2025-11-03 |
| 169 | Grok 4.3 (Non-reasoning) | SpaceXAI | 24.75 | 1,095.96 | 7.6% | 34.1% | 37.4% | 24.7% | -32.3 | 6.5% | 65.8% | 0% | 闭源 | 否 | 2026-04-30 |
| 170 | K-EXAONE | LG AI Research | 24.7* | — | — | — | 35.6% | 55.7% | -57.93 | 13.1% | 78.3% | 1.1% | 开放 | 是 | 2025-12-31 |
| 171 | MiMo-V2-Flash | Xiaomi | 24.67 | 833.05 | 3.1% | 61.8% | 25.9% | 31.3% | -48.48 | 8% | 65.6% | 0% | 开放 | 否 | 2025-12-16 |
| 172 | DeepSeek V3.2历史 | DeepSeek | 24.66* | — | — | — | 38.7% | 39% | -46.73 | 10.5% | 75.1% | 0.9% | 开放 | 否 | 2025-12-01 |
| 173 | Trinity Large Thinking | Arcee AI | 24.47* | 553.84 | — | — | 36.1% | 33% | -44.18 | 14.7% | 75.2% | 0.9% | 开放 | 是 | 2026-04-01 |
| 174 | Qwen3.6 35B A3B | Alibaba | 24.16 | 1,020.43 | 5.2% | 41.6% | 1.3% | 56.7% | -59.83 | 12.5% | 81.7% | 0% | 开放 | 否 | 2026-04-16 |
| 175 | Qwen3 Max历史 | Alibaba | 24.01* | — | — | — | 38.3% | 46.7% | -43.12 | 11.1% | 76.4% | 0% | 闭源 | 否 | 2025-09-23 |
| 176 | Qwen3.5 35B A3B | Alibaba | 23.99 | 796.95 | 4.9% | 40.8% | 29.3% | 55.3% | -62 | 12.8% | 81.9% | 0.6% | 开放 | 否 | 2026-02-24 |
| 177 | gpt-oss-120b (high) | OpenAI | 23.83 | 799.33 | 12% | 26.2% | 38.9% | 50.7% | -50.05 | 18.5% | 78.2% | 1.1% | 开放 | 是 | 2025-08-05 |
| 178 | Gemini 2.5 Flash (Sep)历史 | 23.8* | — | — | — | 40.5% | 64.3% | -34.72 | 12.7% | 79.3% | 0.3% | 闭源 | 是 | 2025-09-25 | |
| 179 | Claude 4.5 Haiku | Anthropic | 23.71* | — | — | — | 34.4% | 43.7% | -7.67 | 4.3% | 64.6% | 0% | 闭源 | 否 | 2025-10-15 |
| 180 | Kimi K2 0905历史 | Kimi | 23.54* | — | — | — | 30.7% | 52.3% | -26.47 | 6.3% | 76.7% | 0% | 开放 | 否 | 2025-09-05 |
| 181 | Claude 3.7 Sonnet历史 | Anthropic | 23.5* | — | — | — | 37.6% | 48.3% | -9.23 | 4.8% | 65.6% | 0% | 闭源 | 否 | 2025-02-24 |
| 182 | o1历史 | OpenAI | 23.44* | — | — | — | 35.8% | 59.3% | -10.55 | 7.7% | 74.7% | 0.3% | 闭源 | 是 | 2024-12-05 |
| 183 | GLM-4.6历史 | Z AI | 22.98* | — | — | — | 33.1% | 26.3% | -31.6 | 5.2% | 63.2% | 0% | 开放 | 否 | 2025-09-30 |
| 184 | EXAONE 4.5 33B | LG AI Research | 22.96* | — | — | — | 28% | 49.3% | -51.48 | 11.6% | 79.4% | 0.3% | 开放 | 是 | 2026-04-09 |
| 185 | GLM-4.7-Flash历史 | Z AI | 22.89* | — | — | — | 33.7% | 35% | -59.33 | 7.1% | 58.1% | 0.3% | 开放 | 是 | 2026-01-19 |
| 186 | Command A+ | Cohere | 22.51 | 714.2 | 5.8% | 22.8% | 37.8% | 46% | -3.98 | 11.4% | 76.1% | 0.3% | 开放 | 是 | 2026-05-20 |
| 187 | Grok 3 mini Reasoning (high)历史 | SpaceXAI | 22.5* | — | — | — | 40.6% | 50.3% | -6.05 | 11.1% | 79.1% | 0.6% | 闭源 | 是 | 2025-02-19 |
| 188 | Grok 4.20 0309历史 | SpaceXAI | 22.48* | — | — | — | 32.2% | 18% | -42.68 | 22.5% | 78.5% | 0% | 闭源 | 否 | 2026-03-10 |
| 189 | DeepSeek V3.2 Speciale历史 | DeepSeek | 22.24* | — | — | — | 44% | 59.3% | -16.48 | 26.1% | 87.1% | 7.4% | 开放 | 是 | 2025-12-01 |
| 190 | Gemma 4 12B | 22* | — | — | — | 38.2% | 55.3% | -51.88 | 14.8% | 75.3% | 0% | 开放 | 是 | 2026-06-03 | |
| 191 | ERNIE 5.0 Thinking Preview | Baidu | 21.92* | — | — | — | 37.5% | 6.7% | -44.13 | 12.7% | 77.7% | 1.4% | 闭源 | 是 | 2025-11-13 |
| 192 | Grok 4.20 0309 v2历史 | SpaceXAI | 21.84* | — | — | — | 32.8% | 17.3% | -44.6 | 24.2% | 77.6% | 0.3% | 闭源 | 否 | 2026-04-07 |
| 193 | Nova 2.0 Pro Preview (medium) | Amazon | 21.77 | 676.22 | 7.6% | 29.6% | 42.7% | 54.3% | -48.05 | 8.9% | 78.5% | 0% | 闭源 | 是 | 2025-11-27 |
| 194 | Gemma 4 31B | 21.77 | 744.22 | 8.5% | 29.2% | 41.1% | 36% | -51.35 | 11.5% | 76.3% | 0% | 开放 | 否 | 2026-04-02 | |
| 195 | Grok Code Fast 1历史 | SpaceXAI | 21.61* | — | — | — | 36.2% | 48.3% | -36 | 7.5% | 72.7% | 0% | 闭源 | 是 | 2025-08-28 |
| 196 | Qwen3.5 9B | Alibaba | 21.44 | 635.58 | 8.2% | 29.2% | 27.5% | 59% | -52.52 | 13.3% | 80.6% | 0.3% | 开放 | 是 | 2026-03-02 |
| 197 | Mercury 2 | Inception | 21.43 | 697.61 | 9.9% | 27.3% | 38.7% | 36.3% | -52.28 | 15.5% | 77% | 0.8% | 闭源 | 是 | 2026-02-20 |
| 198 | DeepSeek V3.1 Terminus历史 | DeepSeek | 21.41* | — | — | — | 32.1% | 43.3% | -41.88 | 8.4% | 75.1% | 0% | 开放 | 否 | 2025-09-22 |
| 199 | DeepSeek V3.2 Exp历史 | DeepSeek | 21.33* | — | — | — | 39.9% | 43% | -47.08 | 8.6% | 73.8% | 1.4% | 开放 | 否 | 2025-09-29 |
| 200 | Nemotron Cascade 2 30B A3B | NVIDIA | 21.25* | — | — | — | 34.8% | 34% | -52.47 | 11.4% | 75.8% | 0.6% | 开放 | 是 | 2026-03-19 |
| 201 | Apriel-v1.5-15B-Thinker历史 | ServiceNow | 21.23* | — | — | — | 34.8% | 20% | -55.78 | 12% | 71.3% | 1.1% | 开放 | 是 | 2025-09-30 |
| 202 | Qwen3 Coder Next | Alibaba | 21.11 | 711.26 | 5.2% | 38.2% | 32.3% | 40% | -60.72 | 9.3% | 73.7% | 0% | 开放 | 否 | 2026-02-03 |
| 203 | DeepSeek V3.1历史 | DeepSeek | 21.05* | — | — | — | 36.7% | 45% | -41.07 | 6.3% | 73.5% | 0% | 开放 | 否 | 2025-08-21 |
| 204 | Nova 2.0 Omni (medium) | Amazon | 20.95* | — | — | — | 36.2% | 53.7% | -57.83 | 6.8% | 76% | 0% | 闭源 | 是 | 2025-11-26 |
| 205 | DeepSeek V3.1历史 | DeepSeek | 20.67* | — | — | — | 39.1% | 53.3% | -28.37 | 13% | 77.9% | 2% | 开放 | 是 | 2025-08-21 |
| 206 | Qwen3 VL 235B A22B历史 | Alibaba | 20.6* | — | — | — | 39.9% | 58.7% | -44.62 | 10.1% | 77.2% | 0% | 开放 | 是 | 2025-09-23 |
| 207 | Apriel-v1.6-15B-Thinker | ServiceNow | 20.54* | — | — | — | 37.3% | 50.3% | -58.53 | 9.8% | 73.3% | 0.3% | 开放 | 是 | 2025-11-25 |
| 208 | GPT-5.1历史 | OpenAI | 20.4* | — | — | — | 36.5% | 44% | -34.95 | 5.2% | 64.3% | 0% | 闭源 | 否 | 2025-11-13 |
| 209 | Qwen3.5 9B | Alibaba | 20.32* | — | 4.1% | 21.3% | 27.7% | 38% | -70.75 | 8.6% | 78.6% | 0.6% | 开放 | 否 | 2026-03-02 |
| 210 | Gemma 4 26B A4B | 20.1* | — | — | — | 37.3% | 39.7% | -62.03 | 10.7% | 71.4% | 0% | 开放 | 否 | 2026-04-02 | |
| 211 | Qwen3.5 4B | Alibaba | 20.09* | — | 8.2% | 25.8% | 16.1% | 55.7% | -56.77 | 7.8% | 77.1% | 0% | 开放 | 是 | 2026-03-02 |
| 212 | DeepSeek R1 0528历史 | DeepSeek | 20.08* | — | — | — | 40.3% | 54.7% | -27.08 | 14.9% | 81.3% | 1.4% | 开放 | 是 | 2025-05-28 |
| 213 | Gemini 2.5 Flash历史 | 20.06* | — | — | — | 39.4% | 61.7% | -27.85 | 11.1% | 79% | 1.1% | 闭源 | 是 | 2025-05-20 | |
| 214 | GPT-5 nano (high)历史 | OpenAI | 19.87* | — | — | — | 36.6% | 41.7% | -27.75 | 8.2% | 67.6% | 0% | 闭源 | 是 | 2025-08-07 |
| 215 | North Mini Code | Cohere | 19.8 | 533.97 | 6.4% | 35.6% | 38.2% | 32.3% | -49.07 | 10% | 75.7% | 0.3% | 开放 | 是 | 2026-06-09 |
| 216 | Qwen3 235B A22B 2507历史 | Alibaba | 19.61 | 540.61 | 7.8% | 12% | 42.4% | 67% | -45.55 | 15% | 79% | 0% | 开放 | 是 | 2025-07-25 |
| 217 | Nova 2.0 Pro Preview (low) | Amazon | 19.61 | 645.19 | 9.1% | 19.5% | 38.7% | 61.7% | -45.77 | 5.2% | 75.1% | 0% | 闭源 | 是 | 2025-11-27 |
| 218 | Mistral Small 4 | Mistral | 19.56 | 587.99 | 5.2% | 21% | 38% | 44.7% | -29.9 | 9.5% | 76.9% | 0.3% | 开放 | 是 | 2026-03-16 |
| 219 | GLM-4.5历史 | Z AI | 19.49* | — | — | — | 34.8% | 48.3% | -25.45 | 12.2% | 78.2% | 0% | 开放 | 是 | 2025-07-28 |
| 220 | Kimi K2历史 | Kimi | 19.4* | — | — | — | 34.5% | 51% | -27.52 | 7% | 76.6% | 0% | 开放 | 否 | 2025-07-11 |
| 221 | GPT-4.1历史 | OpenAI | 19.36* | — | — | — | 38.1% | 61% | -36.18 | 4.6% | 66.6% | 0% | 闭源 | 否 | 2025-04-14 |
| 222 | Qwen3 Max (Preview)历史 | Alibaba | 19.18* | — | — | — | 37% | 39.7% | -40.58 | 9.3% | 76.4% | 0.9% | 闭源 | 否 | 2025-09-05 |
| 223 | Devstral 2 | Mistral | 19.16 | 739.05 | 10.3% | 30.3% | 33.1% | 30% | -46.17 | 3.6% | 59.4% | 0% | 开放 | 否 | 2025-12-09 |
| 224 | GPT-5 nano (medium)历史 | OpenAI | 19* | — | — | — | 33.8% | 40% | -24.88 | 7.6% | 67% | 0% | 闭源 | 是 | 2025-08-07 |
| 225 | Nova 2.0 Lite (medium) | Amazon | 19* | — | — | — | 36.8% | 58.3% | -55.62 | 8.6% | 76.8% | 0% | 闭源 | 是 | 2025-10-29 |
| 226 | Qwen3.5 Omni Flash | Alibaba | 18.99* | — | — | — | 25.5% | 44% | -65.63 | 7.1% | 74.2% | 0% | 闭源 | 否 | 2026-03-30 |
| 227 | Gemini 2.5 Flash (Sep)历史 | 18.83* | — | — | — | 37.5% | 56.7% | -38.12 | 7.8% | 76.6% | 0% | 闭源 | 否 | 2025-09-25 | |
| 228 | JT-MINI | China Mobile | 18.53* | — | — | — | 27.2% | 11.7% | -64.58 | 6.6% | 67.6% | 0% | 闭源 | 否 | 2026-04-15 |
| 229 | DeepSeek R1 (Jan)历史 | DeepSeek | 18.51 | 525.05 | 6.4% | 19.1% | 35.7% | 52.3% | -31.32 | 9.3% | 70.8% | 0.6% | 开放 | 是 | 2025-01-20 |
| 230 | Grok 3历史 | SpaceXAI | 18.35* | — | — | — | 36.8% | 54.7% | -32.33 | 5.1% | 69.3% | 0% | 闭源 | 否 | 2025-02-19 |
| 231 | Seed-OSS-36B-Instruct历史 | ByteDance Seed | 18.34* | — | — | — | 36.5% | 57.7% | -50.85 | 9.1% | 72.6% | 0% | 开放 | 是 | 2025-08-20 |
| 232 | Nova 2.0 Lite (high) | Amazon | 18.17 | 585.55 | 6.8% | 16.1% | 36.9% | 55.3% | -54.12 | 10.9% | 81.1% | 0.3% | 闭源 | 是 | 2025-10-29 |
| 233 | Qwen3 235B 2507历史 | Alibaba | 18.16* | — | — | — | 36% | 31.2% | -42.1 | 10.6% | 75.3% | 0% | 开放 | 否 | 2025-07-21 |
| 234 | Qwen3 Coder 480B历史 | Alibaba | 17.98* | — | — | — | 35.9% | 42.3% | -22.2 | 4.4% | 61.8% | 0% | 开放 | 否 | 2025-07-22 |
| 235 | Qwen3 VL 32B历史 | Alibaba | 17.93* | — | — | — | 28.5% | 55.3% | -50.2 | 9.6% | 73.3% | 0% | 开放 | 是 | 2025-10-21 |
| 236 | Magistral Medium 1.2 | Mistral | 17.87 | 402.93 | 5.6% | 12.4% | 39.2% | 51.3% | -26.23 | 9.6% | 73.9% | 0.3% | 闭源 | 是 | 2025-09-18 |
| 237 | Nova 2.0 Lite (low) | Amazon | 17.81* | — | — | — | 33.3% | 52% | -51.07 | 4.2% | 69.8% | 0% | 闭源 | 是 | 2025-10-29 |
| 238 | HyperNova 60B 2605 | Multiverse Computing | 17.8 | 653.65 | 5.4% | 18.4% | 33% | 31.7% | -56.43 | 15.1% | 73.3% | 0.9% | 开放 | 是 | 2026-05-26 |
| 239 | MiniMax M1 80k历史 | MiniMax | 17.67* | — | — | — | 37.4% | 54.3% | -47.42 | 8.2% | 69.7% | 0% | 开放 | 是 | 2025-06-17 |
| 240 | GPT-5.4 nano历史 | OpenAI | 17.61* | — | — | — | 35.2% | 24.7% | -38.58 | 4.2% | 55.8% | 0% | 闭源 | 否 | 2026-03-17 |
| 241 | Devstral Small 2 | Mistral | 17.44 | 728.83 | 9.7% | 29.6% | 28.8% | 24% | -56.73 | 3.4% | 53.2% | 0% | 开放 | 否 | 2025-12-09 |
| 242 | K2 Think V2 | MBZUAI Institute of Foundation Models | 17.26 | 369.86 | 5.4% | 15% | 33% | 52.7% | -33.92 | 9.5% | 71.3% | 0% | 开放 | 是 | 2025-12-15 |
| 243 | LongCat Flash Lite | LongCat | 17.22* | — | — | — | 28.4% | 25.7% | -70.1 | 6% | 63.6% | 0% | 开放 | 否 | 2026-01-28 |
| 244 | GPT-5 (minimal)历史 | OpenAI | 17.18* | — | — | — | 38.8% | 25% | -33.95 | 5.4% | 67.3% | 0% | 闭源 | 否 | 2025-08-07 |
| 245 | HyperCLOVA X SEED Think (32B) | Naver | 17.02* | — | — | — | 28.4% | 11.7% | -52.87 | 5.5% | 61.5% | 0% | 开放 | 是 | 2025-12-26 |
| 246 | Grok 4.1 Fast历史 | SpaceXAI | 16.88* | — | — | — | 29.6% | 22% | -50.95 | 5% | 63.7% | 0% | 闭源 | 否 | 2025-11-19 |
| 247 | GLM-4.6V历史 | Z AI | 16.75* | — | — | — | 30.4% | 40.3% | -26.48 | 8.9% | 71.9% | 0% | 开放 | 是 | 2025-12-08 |
| 248 | K-EXAONE | LG AI Research | 16.74* | — | — | — | 27% | 47% | -61.17 | 5.4% | 69.5% | 0% | 开放 | 否 | 2025-12-31 |
| 249 | Qwen3 Next 80B A3B | Alibaba | 16.72 | 367.35 | 6.2% | 6.7% | 38.8% | 60.3% | -50.57 | 11.7% | 75.9% | 0% | 开放 | 是 | 2025-09-11 |
| 250 | GPT-5.4 mini历史 | OpenAI | 16.62* | 784.31 | — | — | 39.6% | 30.3% | -45.85 | 5.7% | 60.6% | 0% | 闭源 | 否 | 2026-03-17 |
| 251 | Nova 2.0 Omni (low) | Amazon | 16.56* | — | — | — | 34.3% | 51% | -49.98 | 4% | 69.9% | 0% | 闭源 | 是 | 2025-11-26 |
| 252 | GLM-4.5-Air历史 | Z AI | 16.52* | — | — | — | 30.6% | 43.7% | -62.52 | 6.8% | 73.3% | 0% | 开放 | 是 | 2025-07-28 |
| 253 | Grok 4 Fast历史 | SpaceXAI | 16.47* | — | — | — | 32.9% | 20% | -49.87 | 5% | 60.6% | 0% | 闭源 | 否 | 2025-09-19 |
| 254 | Mi:dm K 2.5 Pro | Korea Telecom | 16.42* | — | — | — | 33.2% | 9% | -57.27 | 7.7% | 70.1% | 0% | 闭源 | 是 | 2025-12-11 |
| 255 | Ring-1T历史 | InclusionAI | 16.16* | — | — | — | 36.7% | 45.7% | -47.87 | 10.2% | 77.4% | 0.6% | 开放 | 是 | 2025-10-13 |
| 256 | Qwen3.5 4B | Alibaba | 16* | — | 3.1% | 21.3% | 18.3% | 28.3% | -74.72 | 7.5% | 71.2% | 0% | 开放 | 否 | 2026-03-02 |
| 257 | Mistral Large 3 | Mistral | 15.93 | 632.51 | 5.8% | 12% | 36.2% | 34.7% | -39.43 | 4.1% | 68% | 0% | 开放 | 否 | 2025-12-02 |
| 258 | o3-mini (high)历史 | OpenAI | 15.64 | 468.11 | 5.2% | 4.5% | 39.8% | 39.3% | -41.33 | 12.3% | 77.3% | 0.3% | 闭源 | 是 | 2025-01-31 |
| 259 | INTELLECT-3 | Prime Intellect | 15.61* | — | — | — | 39.1% | 32.3% | -51.02 | 12.1% | 76.1% | 0.3% | 开放 | 是 | 2025-11-27 |
| 260 | GLM-4.7-Flash历史 | Z AI | 15.52* | — | — | — | 25.5% | 14.7% | -62.23 | 4.9% | 45.2% | 0% | 开放 | 否 | 2026-01-19 |
| 261 | DeepSeek V3 0324历史 | DeepSeek | 15.43 | 314.48 | 4.5% | 13.9% | 35.8% | 41% | -37.57 | 5.2% | 65.5% | 0% | 开放 | 否 | 2025-03-25 |
| 262 | Solar Open 100B | Upstage | 15.15* | — | — | — | 26.9% | 36% | -54.1 | 9.2% | 65.7% | 0% | 开放 | 是 | 2025-12-17 |
| 263 | Gemini 2.5 Flash-Lite (Sep)历史 | 15.13* | — | — | — | 28.7% | 59% | -53.08 | 6.6% | 70.9% | 0% | 闭源 | 是 | 2025-09-08 | |
| 264 | Nemotron 3 Nano Omni 30B A3B Reasoning | NVIDIA | 14.93* | — | — | — | 27.8% | 35.7% | -56.03 | 5.3% | 46.9% | 0% | 开放 | 是 | 2026-04-29 |
| 265 | gpt-oss-120b (low) | OpenAI | 14.91 | 396.55 | 2.9% | 13.9% | 36% | 43.7% | -50.15 | 5.2% | 67.2% | 0% | 开放 | 是 | 2025-08-05 |
| 266 | gpt-oss-20b (high) | OpenAI | 14.89 | 559.03 | 7% | 13.9% | 34.4% | 30.7% | -63.92 | 9.8% | 68.8% | 1.4% | 开放 | 是 | 2025-08-05 |
| 267 | GPT-4.1 mini历史 | OpenAI | 14.77 | 502.92 | 5.2% | 10.1% | 40.4% | 42.3% | -50.13 | 4.6% | 66.4% | 0% | 闭源 | 否 | 2025-04-14 |
| 268 | Mistral Small 3.1历史 | Mistral | 14.74 | 590.15 | 7% | 26.2% | 26.5% | 19.7% | -50.32 | 4.8% | 45.4% | 0% | 开放 | 否 | 2025-03-17 |
| 269 | Mistral Medium 3.1历史 | Mistral | 14.68 | 599.25 | 8.5% | 13.9% | 33.8% | 19.7% | -45.47 | 4.4% | 58.8% | 0% | 闭源 | 否 | 2025-08-12 |
| 270 | Qwen3 30B A3B 2507历史 | Alibaba | 14.44 | 307.72 | 5.4% | 1.5% | 33.3% | 59% | -56.87 | 9.8% | 70.7% | 0.3% | 开放 | 是 | 2025-07-30 |
| 271 | Nova 2.0 Pro Preview | Amazon | 14.36 | 554.73 | 6.8% | 17.2% | 28.1% | 28.3% | -48.15 | 4% | 63.6% | 0% | 闭源 | 否 | 2025-11-27 |
| 272 | gpt-oss-20b (low) | OpenAI | 14.34* | — | — | — | 34% | 31% | -60.13 | 5.1% | 61.1% | 0% | 开放 | 是 | 2025-08-05 |
| 273 | Qwen3 VL 235B A22B历史 | Alibaba | 14.31* | — | — | — | 35.9% | 31.7% | -50.67 | 6.3% | 71.2% | 0% | 开放 | 否 | 2025-09-23 |
| 274 | Llama 4 Maverick | Meta | 14.27 | -16.15 | 3.9% | 7.9% | 33.1% | 46% | -41.77 | 4.8% | 67.1% | 0% | 开放 | 否 | 2025-04-05 |
| 275 | GPT-5 mini (minimal)历史 | OpenAI | 14.25* | — | — | — | 36.9% | 35.7% | -54.12 | 5% | 68.7% | 0% | 闭源 | 否 | 2025-08-07 |
| 276 | K2-V2 (high) | MBZUAI Institute of Foundation Models | 14.19* | — | — | — | 28.6% | 33.3% | -56.7 | 9.8% | 68.1% | 0% | 开放 | 是 | 2025-12-05 |
| 277 | NVIDIA Nemotron 3 Nano | NVIDIA | 14.18 | 483.84 | 6% | 6.7% | 29.6% | 33.7% | -51.63 | 10.2% | 75.7% | 0.9% | 开放 | 是 | 2025-12-15 |
| 278 | DeepSeek V3 (Dec)历史 | DeepSeek | 14.16 | 217.09 | 4.7% | 16.9% | 35.4% | 29% | -41.35 | 3.6% | 55.7% | 0% | 开放 | 否 | 2024-12-26 |
| 279 | Gemini 2.5 Flash历史 | 14.14* | — | — | — | 29.1% | 45.9% | -42.02 | 5.1% | 68.3% | 1.4% | 闭源 | 否 | 2025-05-20 | |
| 280 | Solar Pro 3 | Upstage | 14.14 | 492.55 | 8% | 12% | 24.7% | 27% | -53.78 | 10.1% | 72.4% | 0% | 闭源 | 是 | 2026-04-06 |
| 281 | Ling 2.6 Flash | InclusionAI | 14.05 | 544.69 | 2.9% | 24.3% | 27.1% | 25% | -65.68 | 6.2% | 59.3% | 0% | 开放 | 否 | 2026-04-21 |
| 282 | Qwen3 Next 80B A3B | Alibaba | 13.72* | — | — | — | 30.7% | 51.3% | -59.1 | 7.3% | 73.8% | 0% | 开放 | 否 | 2025-09-11 |
| 283 | Tri-21B-think Preview | Trillion Labs | 13.62* | — | — | — | 17.8% | 14.7% | -55.28 | 5.7% | 53.8% | 0% | 开放 | 是 | 2026-02-10 |
| 284 | Qwen3 Coder 30B A3B历史 | Alibaba | 13.61* | — | — | — | 27.8% | 29% | -50.85 | 4% | 51.6% | 0% | 开放 | 否 | 2025-07-31 |
| 285 | DiffusionGemma 26B A4B | 13.45 | 547.26 | 6.6% | 12.4% | 34.3% | 14.3% | -59.03 | 10.2% | 66.9% | 0.3% | 开放 | 是 | 2026-06-10 | |
| 286 | Qwen3 235B历史 | Alibaba | 13.43* | — | — | — | 39.9% | 0% | -44.58 | 11.7% | 70% | 0% | 开放 | 是 | 2025-04-28 |
| 287 | Qwen3 VL 30B A3B历史 | Alibaba | 13.34* | — | — | — | 28.8% | 40.7% | -57.32 | 8.7% | 72% | 0% | 开放 | 是 | 2025-10-03 |
| 288 | Gemma 4 12B (Non-reasoning) | 13.19* | — | — | — | 29.7% | 30.7% | -52.73 | 6.2% | 66.1% | 0% | 开放 | 否 | 2026-06-03 | |
| 289 | Gemini 2.5 Flash-Lite (Sep)历史 | 13.1* | — | — | — | 28.5% | 48% | -42.67 | 4.6% | 65.1% | 0% | 闭源 | 否 | 2025-09-25 | |
| 290 | Motif-2-12.7B | Motif Technologies | 12.79* | — | — | — | 28.2% | 13% | -61.18 | 8.2% | 69.5% | 0% | 闭源 | 是 | 2025-12-04 |
| 291 | Ling-1T历史 | InclusionAI | 12.75* | — | — | — | 35.2% | 34.7% | -55.48 | 7.2% | 71.9% | 0.6% | 开放 | 否 | 2025-10-08 |
| 292 | Nova Premier | Amazon | 12.73* | — | — | — | 27.9% | 30% | -36.27 | 4.7% | 56.9% | 0% | 闭源 | 否 | 2025-04-30 |
| 293 | Magistral Medium 1历史 | Mistral | 12.51* | — | — | — | 29.7% | 0% | -26.42 | 9.5% | 67.9% | 0.3% | 闭源 | 是 | 2025-06-10 |
| 294 | Gemma 4 E4B | 12.5* | — | — | — | 24.4% | 30.7% | -20.05 | 3.7% | 57.6% | 0.6% | 开放 | 是 | 2026-04-03 | |
| 295 | Mistral Medium 3历史 | Mistral | 12.49* | — | — | — | 33.1% | 28% | -31.47 | 4.3% | 57.8% | 0% | 闭源 | 否 | 2025-05-07 |
| 296 | K2-V2 (medium) | MBZUAI Institute of Foundation Models | 12.43* | — | — | — | 25.2% | 28% | -49.97 | 4.4% | 59.8% | 0% | 开放 | 是 | 2025-12-05 |
| 297 | Llama Nemotron Super 49B v1.5 | NVIDIA | 12.42* | — | — | — | 34.8% | 34% | -45.82 | 6.8% | 74.8% | 0% | 开放 | 是 | 2025-07-25 |
| 298 | Devstral Medium历史 | Mistral | 12.4* | — | — | — | 29.4% | 28.7% | -31.62 | 3.8% | 49.2% | 0% | 闭源 | 否 | 2025-07-10 |
| 299 | Mistral Small 4 | Mistral | 12.37* | — | — | — | 28.1% | 21.3% | -48.37 | 3.7% | 57.1% | 0.3% | 开放 | 否 | 2026-03-16 |
| 300 | Tri-21B-Think | Trillion Labs | 12.36* | — | — | — | 17.4% | 11% | -63.3 | 6.1% | 60.1% | 0.3% | 开放 | 是 | 2026-02-10 |
| 301 | Gemini 2.0 Flash历史 | 12.27* | — | — | — | 33.3% | 28.3% | -42.55 | 5.3% | 62.3% | 0% | 闭源 | 否 | 2025-02-05 | |
| 302 | Claude 3.5 Haiku历史 | Anthropic | 12.26 | 444.21 | 5.8% | 10.1% | 27.4% | 23.3% | -23.18 | 3.5% | 40.8% | 0% | 闭源 | 否 | 2024-10-22 |
| 303 | Llama 3.3 Nemotron Super 49B历史 | NVIDIA | 12.25* | — | — | — | 28.2% | 17% | -58.33 | 6.5% | 64.3% | 0% | 开放 | 是 | 2025-03-18 |
| 304 | MiniCPM5-1B | OpenBMB | 11.96* | — | — | — | 4.4% | 3.7% | -16.62 | 6.5% | 27.8% | 0% | 开放 | 是 | 2026-05-25 |
| 305 | Qwen3 4B 2507历史 | Alibaba | 11.96* | — | — | — | 25.6% | 37.7% | -60.82 | 5.9% | 66.7% | 0% | 开放 | 是 | 2025-08-06 |
| 306 | Sarvam 105B (high) | Sarvam | 11.94* | — | — | — | 26.4% | 0% | -59.53 | 10.1% | 73.8% | 0% | 开放 | 是 | 2026-03-06 |
| 307 | Nova 2.0 Lite | Amazon | 11.83* | — | — | — | 24% | 17.7% | -59.1 | 3% | 60.3% | 0% | 闭源 | 否 | 2025-10-29 |
| 308 | Devstral Small (May)历史 | Mistral | 11.82* | — | — | — | 24.5% | 26.7% | -56.98 | 4% | 43.4% | 0% | 开放 | 否 | 2025-05-21 |
| 309 | MiniCPM5-1B | OpenBMB | 11.74* | — | — | — | 1.4% | 4.7% | -0.78 | 4.6% | 26.9% | 0% | 开放 | 否 | 2026-05-25 |
| 310 | Qwen3 32B历史 | Alibaba | 11.55 | 274.98 | 5.4% | 5.2% | 35.4% | 0% | -50.53 | 8.3% | 66.8% | 0.3% | 开放 | 是 | 2025-04-28 |
| 311 | Gemini 2.5 Flash-Lite历史 | 11.41* | — | — | — | 19.3% | 51.3% | -44.63 | 6.4% | 62.5% | 0% | 闭源 | 是 | 2025-06-17 | |
| 312 | Magistral Small 1.2 | Mistral | 11.29 | 248.36 | 4.3% | 4.5% | 35.2% | 16.3% | -64.98 | 6.1% | 66.3% | 0.3% | 开放 | 是 | 2025-09-17 |
| 313 | GPT-4o (Nov)历史 | OpenAI | 11.18* | — | — | — | 33.3% | 0% | -10.75 | 3.3% | 54.3% | 0% | 闭源 | 否 | 2024-11-20 |
| 314 | Nanbeige4.1-3B | Nanbeige | 11.09 | -141.11 | 0% | 1.1% | 26.6% | 0% | -41.93 | 10% | 84.9% | 0% | 开放 | 是 | 2026-02-11 |
| 315 | Qwen3 VL 32B历史 | Alibaba | 11.06* | — | — | — | 30.1% | 31.3% | -62.82 | 6.3% | 67.1% | 0% | 开放 | 否 | 2025-10-21 |
| 316 | Ministral 3 14B | Mistral | 11.05 | 475.86 | 6.6% | 9.7% | 23.6% | 22% | -66.82 | 4.6% | 57.2% | 0% | 开放 | 否 | 2025-12-02 |
| 317 | GLM-4.6V历史 | Z AI | 10.98* | — | — | — | 27.2% | 12.3% | -37.87 | 3.7% | 56.6% | 0% | 开放 | 否 | 2025-12-08 |
| 318 | Qwen3 235B历史 | Alibaba | 10.86* | — | — | — | 29.9% | 0% | -52.97 | 4.7% | 61.3% | 0% | 开放 | 否 | 2025-04-28 |
| 319 | Magistral Small 1历史 | Mistral | 10.7* | — | — | — | 24.1% | 0% | -43.37 | 7.2% | 64.1% | 0% | 开放 | 是 | 2025-06-10 |
| 320 | EXAONE 4.0 32B | LG AI Research | 10.59* | — | — | — | 34.4% | 14% | -60.8 | 10.5% | 73.9% | 0% | 开放 | 是 | 2025-07-15 |
| 321 | Qwen3 VL 8B历史 | Alibaba | 10.58* | — | — | — | 21.9% | 31% | -52.42 | 3.3% | 57.9% | 0.3% | 开放 | 是 | 2025-10-14 |
| 322 | Mistral Small 3.2历史 | Mistral | 10.56 | 57.17 | 6% | 5.6% | 26.4% | 17.3% | -50.42 | 4.3% | 50.5% | 0% | 开放 | 否 | 2025-06-20 |
| 323 | Nova 2.0 Omni | Amazon | 10.53* | — | — | — | 27.9% | 22.3% | -64.5 | 3.9% | 55.5% | 0% | 闭源 | 否 | 2025-11-26 |
| 324 | Qwen3 14B历史 | Alibaba | 10.38 | 222.45 | 5.4% | 4.9% | 31.6% | 0% | -49.23 | 4.3% | 60.4% | 0% | 开放 | 是 | 2025-04-28 |
| 325 | DeepSeek R1 0528 Qwen3 8B历史 | DeepSeek | 10.37* | — | — | — | 20.4% | 13% | -64.12 | 5.6% | 61.2% | 0% | 开放 | 是 | 2025-05-29 |
| 326 | Llama 4 Scout | Meta | 10.04 | 89.68 | 3.3% | 3.7% | 17% | 25.8% | -52.37 | 4.3% | 58.7% | 0% | 开放 | 否 | 2025-04-05 |
| 327 | Qwen3 VL 30B A3B历史 | Alibaba | 10.02* | — | — | — | 30.8% | 23.7% | -63 | 6.4% | 69.5% | 0% | 开放 | 否 | 2025-10-03 |
| 328 | Hermes 4 70B | Nous Research | 9.97* | — | — | — | 34.1% | 6.7% | -49.92 | 7.9% | 69.9% | 0% | 开放 | 是 | 2025-08-27 |
| 329 | DeepSeek R1 Distill Llama 70B历史 | DeepSeek | 9.93* | — | — | — | 31.3% | 11% | -46.45 | 6.1% | 40.2% | 0% | 开放 | 是 | 2025-01-20 |
| 330 | Falcon-H1R-7B | TII UAE | 9.79* | — | — | — | 24.9% | 8.7% | -62.08 | 10.8% | 66.1% | 0.3% | 开放 | 是 | 2026-01-04 |
| 331 | Ling-flash-2.0历史 | InclusionAI | 9.74* | — | — | — | 28.9% | 15% | -63.37 | 6.3% | 65.7% | 0% | 开放 | 否 | 2025-09-17 |
| 332 | Qwen3 Omni 30B A3B | Alibaba | 9.63* | — | — | — | 30.6% | 0% | -60.72 | 7.3% | 72.6% | 0% | 开放 | 是 | 2025-09-22 |
| 333 | GPT-4o (Aug)历史 | OpenAI | 9.58* | — | — | — | 33.1% | 35% | -21.72 | 2.9% | 52.1% | 0% | 闭源 | 否 | 2024-08-06 |
| 334 | GPT-4.1 nano历史 | OpenAI | 9.58 | 41.4 | 3.5% | 3.7% | 25.9% | 17% | -56.42 | 3.9% | 51.2% | 0% | 闭源 | 否 | 2025-04-14 |
| 335 | Qwen2.5 72B历史 | Alibaba | 9.57* | — | — | — | 26.7% | 20.3% | -52.22 | 4.2% | 49.1% | 0% | 开放 | 否 | 2024-09-19 |
| 336 | Step3 VL 10B | StepFun | 9.47* | — | — | — | 31.1% | 0% | -59 | 10.2% | 69% | 0% | 开放 | 是 | 2026-01-20 |
| 337 | Llama 3.3 70B | Meta | 9.4 | 79.94 | 1% | 4.9% | 26% | 15% | -51.92 | 4% | 49.8% | 0% | 开放 | 否 | 2024-12-06 |
| 338 | Qwen3 30B历史 | Alibaba | 9.31* | — | — | — | 28.5% | 0% | -51.2 | 6.6% | 61.6% | 0% | 开放 | 是 | 2025-04-28 |
| 339 | Gemma 4 E2B | 9.26* | — | — | — | 20.9% | 15% | -23.98 | 4.8% | 43.3% | 0% | 开放 | 是 | 2026-04-02 | |
| 340 | Devstral Small历史 | Mistral | 9.26* | — | — | — | 24.3% | 17% | -51.15 | 3.7% | 41.4% | 0% | 开放 | 否 | 2025-07-10 |
| 341 | Mistral Large 2 (Nov)历史 | Mistral | 9.15* | — | — | — | 29.2% | 5.3% | -34.05 | 4% | 48.6% | 0% | 开放 | 否 | 2024-11-18 |
| 342 | GLM-4.5V历史 | Z AI | 9.15* | — | — | — | 22.1% | 0% | -43.57 | 5.9% | 68.4% | 0% | 开放 | 是 | 2025-08-11 |
| 343 | Llama Nemotron Ultra | NVIDIA | 9.08* | — | — | — | 34.7% | 7.3% | -45.52 | 8.1% | 72.8% | 0% | 开放 | 是 | 2025-04-07 |
| 344 | Qwen3 30B A3B 2507历史 | Alibaba | 9.06* | — | — | — | 30.4% | 22.7% | -65.92 | 6.8% | 65.9% | 0% | 开放 | 否 | 2025-07-29 |
| 345 | ERNIE 4.5 300B A47B | Baidu | 9.02* | — | — | — | 31.5% | 2.3% | -35.88 | 3.5% | 81.1% | 0% | 开放 | 否 | 2025-06-30 |
| 346 | Hermes 4 405B | Nous Research | 9* | — | — | — | 25.2% | 20.7% | -36.05 | 10.3% | 72.7% | 0.3% | 开放 | 是 | 2025-08-27 |
| 347 | Solar Pro 2 | Upstage | 8.99* | — | — | — | 30.2% | 0% | -56.57 | 7% | 68.7% | 0% | 闭源 | 是 | 2025-07-09 |
| 348 | NVIDIA Nemotron Nano 12B v2 VL | NVIDIA | 8.96* | — | — | — | 26.2% | 40% | -64.42 | 5.3% | 57.2% | 0% | 开放 | 是 | 2025-10-28 |
| 349 | Ministral 3 8B | Mistral | 8.96 | 448.72 | 3.7% | 4.1% | 20.8% | 24% | -67.83 | 4.3% | 47.1% | 0% | 开放 | 否 | 2025-12-02 |
| 350 | Gemma 4 E4B | 8.91* | — | — | — | 3.9% | 18% | -41.52 | 4.7% | 54.9% | 0.3% | 开放 | 否 | 2026-04-03 | |
| 351 | Granite 4.1 30B | IBM | 8.86 | 422.01 | 3.9% | 2.6% | 25.8% | 18.7% | -67.78 | 4.2% | 48.1% | 0% | 开放 | 否 | 2026-04-29 |
| 352 | NVIDIA Nemotron Nano 9B V2 | NVIDIA | 8.84* | — | — | — | 22% | 21% | -42.87 | 4.6% | 57% | 0% | 开放 | 是 | 2025-08-18 |
| 353 | Hermes 4 405B | Nous Research | 8.8* | — | — | — | 34.6% | 20% | -33 | 4.2% | 53.6% | 0% | 开放 | 否 | 2025-08-27 |
| 354 | NVIDIA Nemotron 3 Nano 4B | NVIDIA | 8.73 | 192.39 | 7% | 3.7% | 16.4% | 16.7% | -71.53 | 4.8% | 51.3% | 0% | 开放 | 是 | 2026-03-16 |
| 355 | Llama Nemotron Super 49B v1.5 | NVIDIA | 8.68* | — | — | — | 23.8% | 22% | -45.63 | 4.3% | 48.1% | 0% | 开放 | 否 | 2025-07-25 |
| 356 | K2-V2 (low) | MBZUAI Institute of Foundation Models | 8.55* | — | — | — | 22.3% | 19% | -48.07 | 3.9% | 54.1% | 0% | 开放 | 是 | 2025-12-05 |
| 357 | Llama 3.1 405B | Meta | 8.5* | — | — | — | 29.9% | 24.3% | -17.3 | 4.2% | 51.5% | 0% | 开放 | 否 | 2024-07-23 |
| 358 | Llama 3.3 Nemotron Super 49B历史 | NVIDIA | 8.47* | — | — | — | 22.9% | 11.3% | -49.68 | 3.5% | 51.7% | 0% | 开放 | 否 | 2025-03-18 |
| 359 | Qwen3 VL 8B历史 | Alibaba | 8.42* | — | — | — | 17.4% | 15.3% | -52.05 | 2.9% | 42.7% | 0% | 开放 | 否 | 2025-10-14 |
| 360 | LFM2.5-8B-A1B | Liquid AI | 8.32* | — | — | — | 7.8% | 0% | -33.27 | 6.9% | 51.3% | 0% | 开放 | 是 | 2026-05-28 |
| 361 | Qwen3 8B历史 | Alibaba | 8.29 | 200.34 | 4.5% | 2.2% | 22.6% | 0% | -65.07 | 4.2% | 58.9% | 0% | 开放 | 是 | 2025-04-28 |
| 362 | Ring-flash-2.0 | InclusionAI | 8.16* | — | — | — | 16.8% | 21% | -58.67 | 8.9% | 72.5% | 0.3% | 开放 | 是 | 2025-09-19 |
| 363 | Olmo 3.1 32B Think | Allen Institute for AI | 8.09* | — | — | — | 29.3% | 0% | -44.12 | 6% | 59.1% | 0% | 开放 | 是 | 2025-12-12 |
| 364 | GPT-5 nano (minimal)历史 | OpenAI | 8* | — | — | — | 29.1% | 20% | -63.98 | 4.1% | 42.8% | 0% | 闭源 | 否 | 2025-08-07 |
| 365 | Qwen3 VL 4B历史 | Alibaba | 7.9* | — | — | — | 17.1% | 21.3% | -69.35 | 4.4% | 49.4% | 0% | 开放 | 是 | 2025-10-14 |
| 366 | Solar Pro 2 | Upstage | 7.77* | — | — | — | 24.8% | 0% | -61.68 | 3.8% | 56.1% | 0% | 闭源 | 否 | 2025-07-09 |
| 367 | Nova Pro历史 | Amazon | 7.67* | — | — | — | 20.8% | 19% | -47.58 | 3.4% | 49.9% | 0% | 闭源 | 否 | 2024-12-03 |
| 368 | Command A | Cohere | 7.67* | — | — | — | 28.1% | 18% | -48.47 | 4.6% | 52.7% | 0% | 开放 | 否 | 2025-03-13 |
| 369 | Llama 3.1 Nemotron 70B | NVIDIA | 7.64* | — | — | — | 23.3% | 7% | -41.05 | 4.6% | 46.5% | 0% | 开放 | 否 | 2024-10-15 |
| 370 | Qwen3.5 2B | Alibaba | 7.64 | 198.86 | 2.5% | 3% | 2.8% | 23.7% | -47.48 | 2.1% | 45.6% | 0% | 开放 | 是 | 2026-03-02 |
| 371 | Llama 3.1 8B历史 | Meta | 7.61 | -123.73 | 1.6% | 1.5% | 13.2% | 15.7% | -30.33 | 5.1% | 25.9% | 0% | 开放 | 否 | 2024-07-23 |
| 372 | Gemma 3 27B历史 | 7.42 | -144.46 | 0.8% | 4.5% | 21.2% | 5.7% | -65.88 | 4.7% | 42.8% | 0% | 开放 | 否 | 2025-03-12 | |
| 373 | NVIDIA Nemotron 3 Nano | NVIDIA | 7.39* | -94.34 | — | — | 23% | 6.7% | -69.22 | 4.6% | 39.9% | 0% | 开放 | 否 | 2025-12-15 |
| 374 | NVIDIA Nemotron Nano 9B V2 | NVIDIA | 7.38* | — | — | — | 20.9% | 22.7% | -57.1 | 4% | 55.7% | 0% | 开放 | 否 | 2025-08-18 |
| 375 | Qwen3 4B 2507历史 | Alibaba | 7.12* | — | — | — | 18.1% | 7.3% | -59.63 | 4.7% | 51.7% | 0% | 开放 | 否 | 2025-08-06 |
| 376 | Qwen3 14B历史 | Alibaba | 7.02* | — | — | — | 26.5% | 0% | -67.38 | 4.2% | 47% | 0% | 开放 | 否 | 2025-04-28 |
| 377 | GLM-4.5V历史 | Z AI | 7* | — | — | — | 18.8% | 0% | -53.52 | 3.6% | 57.3% | 0% | 开放 | 否 | 2025-08-11 |
| 378 | Nova Lite历史 | Amazon | 6.92* | — | — | — | 13.9% | 17.7% | -40.42 | 4.6% | 43.3% | 0% | 闭源 | 否 | 2024-12-03 |
| 379 | Gemini 2.5 Flash-Lite历史 | 6.92* | — | — | — | 17.7% | 31.3% | -57 | 3.7% | 47.4% | 0% | 闭源 | 否 | 2025-06-17 | |
| 380 | Hermes 4 70B | Nous Research | 6.9* | — | — | — | 27.7% | 2% | -47.37 | 3.6% | 49.1% | 0% | 开放 | 否 | 2025-08-27 |
| 381 | Ministral 3 3B | Mistral | 6.8 | 272.8 | 4.7% | 0% | 14.4% | 11.7% | -64.1 | 5.3% | 35.8% | 0% | 开放 | 否 | 2025-12-02 |
| 382 | Qwen3 30B历史 | Alibaba | 6.8* | — | — | — | 26.4% | 0% | -65.07 | 4.6% | 51.5% | 0% | 开放 | 否 | 2025-04-28 |
| 383 | Llama 3.1 70B历史 | Meta | 6.75* | — | — | — | 26.7% | 6.3% | -40.85 | 4.6% | 40.9% | 0% | 开放 | 否 | 2024-07-23 |
| 384 | Granite 4.1 8B | IBM | 6.67* | — | 3.1% | 3.4% | 21.8% | 12% | -64.6 | 3.8% | 43.3% | 0% | 开放 | 否 | 2026-04-29 |
| 385 | Sarvam 30B (high) | Sarvam | 6.64* | — | — | — | 19.2% | 0% | -72.05 | 7% | 63.3% | 0.3% | 开放 | 是 | 2026-03-06 |
| 386 | Olmo 3.1 32B Instruct | Allen Institute for AI | 6.47* | — | — | — | 16.7% | 0% | -51.08 | 4.9% | 53.9% | 0% | 开放 | 否 | 2026-01-13 |
| 387 | Gemma 4 E2B | 6.41* | — | — | — | 20.4% | 15% | -61.37 | 4.5% | 40.5% | 0% | 开放 | 否 | 2026-04-02 | |
| 388 | Olmo 3 32B Think历史 | Allen Institute for AI | 6.4* | — | — | — | 28.6% | 0% | -58.4 | 5.9% | 61% | 0% | 开放 | 是 | 2025-11-20 |
| 389 | Phi-4 Mini | Microsoft | 6.01 | -145.05 | 1% | 0.4% | 10.8% | 13.7% | -61.02 | 4.2% | 33.1% | 0% | 开放 | 否 | 2024-02-26 |
| 390 | EXAONE 4.0 32B | LG AI Research | 6.01* | — | — | — | 25.2% | 8% | -62.3 | 4.9% | 62.8% | 0% | 开放 | 否 | 2025-07-15 |
| 391 | Qwen3.5 2B | Alibaba | 5.62 | -103.62 | 2.3% | 0% | 7.2% | 13.7% | -82.83 | 4.9% | 43.8% | 0% | 开放 | 否 | 2026-03-02 |
| 392 | Gemma 3 12B历史 | 5.51 | -143.64 | 0.8% | 0% | 17.4% | 6.7% | -76.98 | 4.8% | 34.9% | 0% | 开放 | 否 | 2025-03-12 | |
| 393 | Qwen3.5 0.8B | Alibaba | 5.48 | -111.26 | 0% | 0% | 0% | 5.3% | -35.18 | 1.2% | 11.4% | 0% | 开放 | 是 | 2026-03-02 |
| 394 | Jamba 1.7 Large | AI21 Labs | 5.3* | — | — | — | 18.8% | 17.3% | -55.55 | 3.8% | 39% | 0% | 开放 | 否 | 2025-07-07 |
| 395 | Granite 4.0 H Small | IBM | 5.24* | — | — | — | 20.9% | 9% | -60.92 | 3.7% | 41.6% | 0% | 开放 | 否 | 2025-09-22 |
| 396 | Qwen3 Omni 30B A3B | Alibaba | 5.11* | — | — | — | 18.6% | 0% | -69.18 | 5.1% | 62% | 0% | 开放 | 否 | 2025-09-22 |
| 397 | Qwen3 8B历史 | Alibaba | 5.07* | — | — | — | 16.8% | 0% | -74.1 | 2.8% | 45.2% | 0% | 开放 | 否 | 2025-04-28 |
| 398 | LFM2 24B A2B | Liquid AI | 4.95* | — | — | — | 10.9% | 0% | -59.07 | 4.4% | 47.4% | 0% | 开放 | 否 | 2026-02-25 |
| 399 | Phi-4 | Microsoft | 4.87* | — | — | — | 26% | 0% | -56.72 | 4.1% | 57.5% | 0% | 开放 | 否 | 2024-12-12 |
| 400 | Nova Micro | Amazon | 4.74* | — | — | — | 9.4% | 9.7% | -48.67 | 4.7% | 35.8% | 0% | 闭源 | 否 | 2024-12-03 |
| 401 | Granite 4.1 3B | IBM | 4.7 | 103.94 | 1.2% | 1.1% | 11.9% | 3% | -77.38 | 3.4% | 31.4% | 0% | 开放 | 否 | 2026-04-29 |
| 402 | NVIDIA Nemotron Nano 12B v2 VL | NVIDIA | 4.58* | — | — | — | 17.6% | 17% | -72.75 | 4.5% | 43.9% | 0% | 开放 | 否 | 2025-10-28 |
| 403 | MiniCPM-V 4.6 1.3B | OpenBMB | 4.19 | -108.35 | 4.3% | 0% | 2.1% | 6.3% | -84.67 | 4.9% | 30.5% | 0% | 开放 | 否 | 2026-05-11 |
| 404 | Jamba Reasoning 3B | AI21 Labs | 4.13* | — | — | — | 5.9% | 7% | -73.68 | 4.6% | 33.3% | 0% | 开放 | 是 | 2025-10-08 |
| 405 | Qwen3 VL 4B历史 | Alibaba | 4.09* | — | — | — | 13.7% | 13% | -75.82 | 3.7% | 37.1% | 0% | 开放 | 否 | 2025-10-14 |
| 406 | Reka Flash 3 | Reka AI | 4.06* | — | — | — | 26.7% | 0% | -64.7 | 5.1% | 52.9% | 0% | 开放 | 是 | 2025-03-10 |
| 407 | Olmo 3 7B Think | Allen Institute for AI | 3.97* | — | — | — | 21.2% | 0% | -73.68 | 5.7% | 51.6% | 0% | 开放 | 是 | 2025-11-20 |
| 408 | Claude 3 Haiku历史 | Anthropic | 3.86* | — | — | — | 18.6% | 21% | -47.65 | 3.9% | 37.4% | 0% | 闭源 | 否 | 2024-03-04 |
| 409 | Ling-mini-2.0 | InclusionAI | 3.75* | — | — | — | 13.5% | 6.7% | -78.77 | 5% | 56.2% | 0% | 开放 | 否 | 2025-09-09 |
| 410 | Llama 3 70B历史 | Meta | 3.47* | — | — | — | 18.9% | 0% | -53.58 | 4.4% | 37.9% | 0% | 开放 | 否 | 2024-04-18 |
| 411 | Llama 3.2 11B (Vision) | Meta | 3.33* | — | — | — | 11.2% | 11.7% | -63.08 | 5.2% | 22.1% | 0% | 开放 | 否 | 2024-09-25 |
| 412 | Qwen3.5 0.8B | Alibaba | 3.31 | -97.24 | 1.2% | 0.4% | 2.9% | 6.7% | -89 | 4.9% | 23.6% | 0% | 开放 | 否 | 2026-03-02 |
| 413 | Sarvam M历史 | Sarvam | 3.03* | — | — | — | 17.8% | 0% | -62.25 | 3.3% | 41.6% | 0% | 开放 | 是 | 2025-05-23 |
| 414 | Exaone 4.0 1.2B | LG AI Research | 2.9* | — | — | — | 9.3% | 0% | -82.43 | 5.8% | 51.5% | 0% | 开放 | 是 | 2025-07-15 |
| 415 | Olmo 3 7B | Allen Institute for AI | 2.81* | — | — | — | 10.3% | 0% | -77.75 | 5.8% | 40% | 0% | 开放 | 否 | 2025-11-20 |
| 416 | Exaone 4.0 1.2B | LG AI Research | 2.77* | — | — | — | 7.4% | 0% | -82.58 | 5.8% | 42.4% | 0% | 开放 | 否 | 2025-07-15 |
| 417 | LFM2.5-1.2B-Thinking | Liquid AI | 2.75* | — | — | — | 4.2% | 0% | -83.87 | 6.1% | 33.9% | 0% | 开放 | 是 | 2026-01-20 |
| 418 | Jamba 1.7 Mini | AI21 Labs | 2.73* | — | — | — | 9.3% | 12.7% | -74.25 | 4.5% | 32.2% | 0% | 开放 | 否 | 2025-07-07 |
| 419 | LFM2 2.6B | Liquid AI | 2.71* | — | — | — | 2.5% | 0% | -52.1 | 5.2% | 29.2% | 0% | 开放 | 否 | 2025-09-23 |
| 420 | LFM2.5-1.2B-Instruct | Liquid AI | 2.71* | — | — | — | 2.3% | 0% | -73.78 | 6.8% | 32.6% | 0% | 开放 | 否 | 2026-01-05 |
| 421 | Granite 4.0 H 1B | IBM | 2.66* | — | — | — | 8.2% | 6.3% | -73.65 | 5% | 26.3% | 0% | 开放 | 否 | 2025-10-28 |
| 422 | Qwen3 1.7B历史 | Alibaba | 2.63* | — | — | — | 4.3% | 0% | -77.2 | 4.8% | 35.6% | 0% | 开放 | 是 | 2025-04-28 |
| 423 | Gemma 3 270M | 2.41* | — | — | — | 0% | 0% | -31.32 | 4.2% | 22.4% | 0% | 开放 | 否 | 2025-08-14 | |
| 424 | Apertus 70B Instruct | Swiss AI Initiative | 2.4* | — | — | — | 5.7% | 0% | -55.1 | 5.5% | 27.2% | 0% | 开放 | 否 | 2025-09-02 |
| 425 | Granite 4.0 Micro | IBM | 2.37* | — | — | — | 11.9% | 4% | -77.92 | 5.1% | 33.6% | 0% | 开放 | 否 | 2025-09-22 |
| 426 | Mistral 7B历史 | Mistral | 2.14* | — | — | — | 2.4% | 0% | -64.77 | 4.3% | 17.7% | 0% | 开放 | 否 | 2023-09-27 |
| 427 | Granite 4.0 1B | IBM | 2.07* | — | — | — | 8.7% | 4% | -81.82 | 5.1% | 28.1% | 0% | 开放 | 否 | 2025-10-28 |
| 428 | Molmo2-8B | Allen Institute for AI | 2.03* | — | — | — | 13.3% | 0% | -69.12 | 4.4% | 42.5% | 0% | 开放 | 否 | 2025-12-11 |
| 429 | LFM2 8B A1B | Liquid AI | 1.78* | — | — | — | 6.8% | 0% | -75.13 | 4.9% | 34.4% | 0% | 开放 | 否 | 2025-10-07 |
| 430 | Granite 3.3 8B历史 | IBM | 1.75* | — | — | — | 10.1% | 4.3% | -76.62 | 4.2% | 33.8% | 0% | 开放 | 否 | 2025-04-16 |
| 431 | Qwen3 1.7B历史 | Alibaba | 1.53* | — | — | — | 6.9% | 0% | -83.47 | 5.2% | 28.3% | 0% | 开放 | 否 | 2025-04-28 |
| 432 | Qwen3 0.6B历史 | Alibaba | 1.28* | — | — | — | 2.8% | 0% | -81.78 | 5.7% | 23.9% | 0% | 开放 | 是 | 2025-04-28 |
| 433 | Llama 3 8B历史 | Meta | 1.19* | — | — | — | 11.9% | 0% | -69.55 | 5.1% | 29.6% | 0% | 开放 | 否 | 2024-04-18 |
| 434 | Gemma 3n E4B历史 | 1.19* | — | 0.2% | 0.7% | 8.1% | 0% | -80.4 | 4.4% | 29.6% | 0% | 开放 | 否 | 2025-06-26 | |
| 435 | LFM2 1.2B历史 | Liquid AI | 1.15* | — | — | — | 2.5% | 0% | -71.62 | 5.7% | 22.8% | 0% | 开放 | 否 | 2025-07-10 |
| 436 | Gemma 3 4B历史 | 1.12* | — | 0.4% | 0.4% | 7.3% | 5.7% | -83.32 | 5.2% | 29.1% | 0% | 开放 | 否 | 2025-03-12 | |
| 437 | Llama 3.2 1B历史 | Meta | 1.1* | — | — | — | 1.7% | 5% | -55.22 | 5.3% | 17.7% | 0% | 开放 | 否 | 2024-09-25 |
| 438 | LFM2.5-VL-1.6B | Liquid AI | 1.01* | — | — | — | 3% | 0% | -83.87 | 5.1% | 28.9% | 0% | 开放 | 否 | 2026-01-05 |
| 439 | Granite 4.0 350M | IBM | 1* | — | — | — | 0.9% | 0% | -72.15 | 5.7% | 20.3% | 0% | 开放 | 否 | 2025-10-28 |
| 440 | Apertus 8B Instruct | Swiss AI Initiative | 1* | — | — | — | 4.1% | 0% | -74.72 | 5% | 25.6% | 0% | 开放 | 否 | 2025-09-02 |
| 441 | Gemma 3 1B历史 | 1* | — | — | — | 0.7% | 0% | -77.27 | 5.2% | 23.7% | 0% | 开放 | 否 | 2025-03-13 | |
| 442 | Gemma 3n E2B历史 | 1* | — | — | — | 5.2% | 0% | -79.58 | 4% | 22.9% | 0% | 开放 | 否 | 2025-06-26 | |
| 443 | Tiny Aya Global | Cohere | 1* | — | — | — | 3.6% | 0% | -84.72 | 5.2% | 30.5% | 0% | 开放 | 否 | 2026-02-17 |
| 444 | Qwen3 0.6B历史 | Alibaba | 1* | — | — | — | 4.1% | 0% | -85.53 | 5.2% | 23.1% | 0% | 开放 | 否 | 2025-04-28 |
| 445 | Granite 4.0 H 350M | IBM | 1* | — | — | — | 1.7% | 0% | -87.25 | 6.4% | 25.7% | 0% | 开放 | 否 | 2025-10-28 |
| 446 | Cogito v2.1 | Deep Cogito | — | — | — | — | 41% | 21.7% | -24.98 | 11% | 76.8% | 0% | 开放 | 是 | 2025-11-18 |
| 447 | Mi:dm K 2.5 Pro Preview | Korea Telecom | — | — | — | — | 29.7% | 11% | -62.02 | 8.8% | 72.2% | 0% | 闭源 | 是 | 2025-12-11 |
拆解模型在知识、推理、编程、工具使用和长上下文上的差异。
评测结果受任务集合、评分方法和模型版本影响。本站保留原始口径,不用它生成新的综合排名。
用于回答一个明确的问题:模型在这一类任务上的相对表现如何,而不是“哪个模型绝对最好”。