模型排行
285 個模型 · 2 個榜單 · 2 個數據來源
人氣王座
開源之光
國產之光
最多人測
開源前三甲
deepseek-r1DeepSeek · 總第 4 名1,317
gemma-3-27b-itGoogle · 總第 12 名1,279
deepseek-v3DeepSeek · 總第 13 名1,277
這一期的變化
QwQ 32B智力榜 · 第 4新上榜
Reka Flash 3智力榜 · 第 12新上榜
Command A智力榜 · 第 26新上榜
Gemma 3 27B Instruct智力榜 · 第 30新上榜
| 排名 | 變化 | 模型 | 得分 | 置信區間 | 投票 | 價格 $/M |
|---|---|---|---|---|---|---|
| 1 | — | deepseek-r1開源DeepSeek | 1,317 | ±5 | 12,224 | — |
| 2 | ▲ 1 | gemma-3-27b-it開源Google | 1,279 | ±7 | 6,248 | — |
| 3 | ▼ 1 | deepseek-v3開源DeepSeek | 1,277 | ±4 | 22,848 | — |
| 4 | 新 | command-a-03-2025開源Cohere | 1,271 | ±10 | 3,000 | — |
| 5 | — | llama-3.1-405b-instruct-bf16開源Meta | 1,253 | ±4 | 39,003 | — |
| 6 | — | llama-3.1-405b-instruct-fp8開源Meta | 1,253 | ±3 | 63,070 | — |
| 7 | ▼ 3 | qwq-32b開源Alibaba · 另有 1 個檔位 | 1,251 | ±9 | 3,403 | — |
| 8 | 新 | llama-3.3-nemotron-49b-super-v1開源NVIDIA | 1,244 | ±12 | 2,232 | — |
| 9 | ▼ 2 | deepseek-v2.5-1210開源DeepSeek · 另有 1 個檔位 | 1,240 | ±6 | 7,252 | — |
| 10 | ▼ 2 | llama-3.3-70b-instruct開源Meta | 1,235 | ±4 | 33,298 | — |
| 11 | ▼ 2 | mistral-large-2407開源Mistral · 另有 1 個檔位 | 1,232 | ±3 | 48,228 | — |
| 12 | ▼ 2 | qwen-max-0919開源Alibaba | 1,231 | ±4 | 17,434 | — |
| 13 | ▼ 2 | athene-v2-chat開源NexusFlow | 1,231 | ±4 | 26,104 | — |
| 14 | ▼ 2 | athene-70b-0725開源NexusFlow | 1,223 | ±4 | 20,595 | — |
| 15 | ▼ 2 | qwen2.5-72b-instruct開源Alibaba | 1,222 | ±3 | 41,549 | — |
| 16 | ▼ 2 | llama-3.1-nemotron-70b-instruct開源NVIDIA | 1,216 | ±7 | 7,579 | — |
| 17 | ▼ 2 | llama-3.1-70b-instruct開源Meta | 1,214 | ±3 | 58,685 | — |
| 18 | ▼ 2 | llama-3.1-tulu-3-70b開源Ai2 | 1,209 | ±9 | 3,014 | — |
| 19 | ▼ 2 | gemma-2-27b-it開源Google | 1,207 | ±3 | 79,531 | — |
| 20 | ▼ 2 | jamba-1.5-large開源AI21 Labs | 1,205 | ±5 | 9,128 | — |
| 21 | ▼ 2 | llama-3.1-nemotron-51b-instruct開源NVIDIA | 1,204 | ±8 | 3,885 | — |
| 22 | ▼ 2 | gemma-2-9b-it-simpo開源Princeton | 1,200 | ±6 | 10,553 | — |
| 23 | ▼ 2 | command-r-plus-08-2024開源Cohere | 1,197 | ±6 | 10,543 | — |
| 24 | ▼ 2 | nemotron-4-340b-instruct開源NVIDIA | 1,196 | ±4 | 20,606 | — |
| 25 | ▼ 2 | llama-3-70b-instruct開源Meta | 1,195 | ±2 | 163,698 | — |
| 26 | ▼ 2 | qwen2.5-coder-32b-instruct開源Alibaba | 1,190 | ±8 | 5,733 | — |
| 27 | ▼ 2 | mistral-small-24b-instruct-2501開源Mistral | 1,187 | ±5 | 12,068 | — |
| 28 | ▼ 2 | c4ai-aya-expanse-32b開源Cohere | 1,185 | ±3 | 28,773 | — |
| 29 | ▼ 2 | command-r-plus開源Cohere | 1,183 | ±2 | 80,874 | — |
| 30 | ▼ 2 | deepseek-coder-v2開源DeepSeek | 1,182 | ±4 | 15,754 | — |
| 31 | ▼ 2 | qwen2-72b-instruct開源Alibaba | 1,181 | ±4 | 38,881 | — |
| 32 | ▼ 2 | gemma-2-9b-it開源Google | 1,180 | ±3 | 57,205 | — |
| 33 | ▼ 2 | command-r-08-2024開源Cohere | 1,176 | ±4 | 10,849 | — |
| 34 | ▼ 2 | phi-4開源Microsoft | 1,176 | ±5 | 20,819 | — |
| 35 | ▼ 2 | jamba-1.5-mini開源AI21 Labs | 1,160 | ±6 | 9,275 | — |
| 36 | ▼ 2 | ministral-8b-2410開源Mistral | 1,160 | ±7 | 5,115 | — |
| 37 | ▼ 2 | qwen1.5-110b-chat開源Alibaba | 1,155 | ±3 | 27,442 | — |
| 38 | ▼ 2 | c4ai-aya-expanse-8b開源Cohere | 1,151 | ±6 | 10,397 | — |
| 39 | ▼ 2 | mixtral-8x22b-instruct-v0.1開源Mistral | 1,149 | ±3 | 53,775 | — |
| 40 | ▼ 2 | qwen1.5-72b-chat開源Alibaba | 1,148 | ±3 | 40,662 | — |
| 41 | ▼ 2 | command-r開源Cohere | 1,148 | ±3 | 56,390 | — |
| 42 | ▼ 2 | llama-3.1-tulu-3-8b開源Ai2 | 1,148 | ±10 | 3,075 | — |
| 43 | ▼ 2 | llama-3-8b-instruct開源Meta | 1,143 | ±3 | 109,111 | — |
| 44 | ▼ 2 | granite-3.1-8b-instruct開源IBM | 1,135 | ±9 | 3,300 | — |
| 45 | ▼ 2 | yi-1.5-34b-chat開源01 AI | 1,133 | ±3 | 25,140 | — |
| 46 | ▼ 2 | llama-3.1-8b-instruct開源Meta | 1,133 | ±3 | 52,608 | — |
| 47 | ▼ 2 | zephyr-orpo-141b-A35b-v0.1開源HuggingFace | 1,130 | ±8 | 4,864 | — |
| 48 | ▼ 2 | qwen1.5-32b-chat開源Alibaba | 1,126 | ±4 | 22,770 | — |
| 49 | ▼ 1 | gemma-2-2b-it開源Google | 1,118 | ±4 | 48,921 | — |
| 50 | ▼ 3 | phi-3-medium-4k-instruct開源Microsoft | 1,118 | ±4 | 26,112 | — |
智力進步
漲得最多 · Anthropic +57.2目前最高 · Anthropic 64.6最高與最低相差 18.1 分時間範圍 2025年11月 – 2026年11月
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi