模型排行
347 个模型 · 2 个榜单 · 2 个数据来源
人气王座
开源之光
国产之光
最多人测
得分前三甲
o3-2025-04-16OpenAI1,435
chatgpt-4o-latest-20250326OpenAI1,422
gpt-4.5-preview-2025-02-27OpenAI1,417
开源前三甲
deepseek-v3-0324DeepSeek · 总第 6 名1,382
deepseek-r1DeepSeek · 总第 9 名1,376
qwen3-235b-a22bAlibaba · 总第 20 名1,343
这一期的变化
Qwen2.5 Turbo智力榜 · 第 70新上榜
| 排名 | 变化 | 模型 | 得分 | 置信区间 | 投票 | 价格 $/M |
|---|---|---|---|---|---|---|
| 1 | — | deepseek-v3-0324开源DeepSeek · 另有 1 个档位 | 1,382 | ±5 | 9,741 | — |
| 2 | — | deepseek-r1开源DeepSeek | 1,376 | ±4 | 19,339 | — |
| 3 | — | qwen3-235b-a22b开源Alibaba | 1,343 | ±9 | 4,942 | — |
| 4 | — | gemma-3-27b-it开源Google | 1,343 | ±5 | 12,989 | — |
| 5 | 新 | qwen3-32b开源Alibaba | 1,328 | ±9 | 3,960 | — |
| 6 | 新 | llama-3.1-nemotron-ultra-253b-v1开源NVIDIA | 1,326 | ±11 | 2,658 | — |
| 7 | ▼ 2 | command-a-03-2025开源Cohere | 1,325 | ±5 | 11,530 | — |
| 8 | ▼ 2 | gemma-3-12b-it开源Google | 1,323 | ±8 | 3,882 | — |
| 9 | ▼ 2 | qwq-32b开源Alibaba · 另有 1 个档位 | 1,316 | ±5 | 9,936 | — |
| 10 | ▼ 2 | llama-3.1-405b-instruct-bf16开源Meta | 1,314 | ±2 | 43,790 | — |
| 11 | ▼ 2 | llama-3.1-405b-instruct-fp8开源Meta | 1,313 | ±2 | 63,035 | — |
| 12 | ▼ 2 | llama-4-maverick-17b-128e-instruct开源Meta | 1,309 | ±7 | 8,797 | — |
| 13 | ▼ 2 | llama-3.3-nemotron-49b-super-v1开源NVIDIA | 1,308 | ±11 | 2,368 | — |
| 14 | ▼ 2 | deepseek-v2.5-1210开源DeepSeek · 另有 1 个档位 | 1,301 | ±7 | 7,245 | — |
| 15 | 新 | qwen3-30b-a3b开源Alibaba | 1,295 | ±9 | 4,159 | — |
| 16 | ▼ 3 | llama-3.3-70b-instruct开源Meta | 1,295 | ±3 | 38,078 | — |
| 17 | 新 | gemma-3n-e4b-it开源Google | 1,293 | ±10 | 3,583 | — |
| 18 | ▼ 4 | mistral-large-2407开源Mistral · 另有 1 个档位 | 1,292 | ±2 | 48,218 | — |
| 19 | ▼ 4 | qwen-max-0919开源Alibaba | 1,292 | ±5 | 17,431 | — |
| 20 | ▼ 4 | athene-v2-chat开源NexusFlow | 1,291 | ±3 | 26,074 | — |
| 21 | 新 | llama-4-scout-17b-16e-instruct开源Meta | 1,290 | ±10 | 3,397 | — |
| 22 | ▼ 5 | athene-70b-0725开源NexusFlow | 1,283 | ±4 | 20,581 | — |
| 23 | ▼ 5 | qwen2.5-72b-instruct开源Alibaba | 1,283 | ±2 | 41,520 | — |
| 24 | ▼ 5 | gemma-3-4b-it开源Google | 1,282 | ±8 | 4,206 | — |
| 25 | 新 | mistral-small-3.1-24b-instruct-2503开源Mistral | 1,279 | ±10 | 2,352 | — |
| 26 | ▼ 6 | llama-3.1-nemotron-70b-instruct开源NVIDIA | 1,277 | ±7 | 7,579 | — |
| 27 | ▼ 6 | llama-3.1-70b-instruct开源Meta | 1,274 | ±2 | 58,638 | — |
| 28 | ▼ 6 | llama-3.1-tulu-3-70b开源Ai2 | 1,269 | ±11 | 3,012 | — |
| 29 | ▼ 6 | gemma-2-27b-it开源Google | 1,267 | ±2 | 79,529 | — |
| 30 | ▼ 6 | jamba-1.5-large开源AI21 Labs | 1,266 | ±6 | 9,125 | — |
| 31 | ▼ 6 | llama-3.1-nemotron-51b-instruct开源NVIDIA | 1,264 | ±8 | 3,888 | — |
| 32 | ▼ 6 | gemma-2-9b-it-simpo开源Princeton | 1,260 | ±5 | 10,547 | — |
| 33 | ▼ 6 | command-r-plus-08-2024开源Cohere | 1,257 | ±5 | 10,536 | — |
| 34 | ▼ 6 | nemotron-4-340b-instruct开源NVIDIA | 1,256 | ±4 | 20,610 | — |
| 35 | ▼ 6 | llama-3-70b-instruct开源Meta | 1,255 | ±2 | 163,623 | — |
| 36 | ▼ 6 | mistral-small-24b-instruct-2501开源Mistral | 1,252 | ±4 | 15,322 | — |
| 37 | ▼ 6 | qwen2.5-coder-32b-instruct开源Alibaba | 1,250 | ±7 | 5,731 | — |
| 38 | ▼ 6 | c4ai-aya-expanse-32b开源Cohere | 1,245 | ±4 | 28,760 | — |
| 39 | ▼ 6 | command-r-plus开源Cohere | 1,243 | ±2 | 80,841 | — |
| 40 | ▼ 6 | deepseek-coder-v2开源DeepSeek | 1,242 | ±4 | 15,753 | — |
| 41 | ▼ 6 | gemma-2-9b-it开源Google | 1,241 | ±2 | 57,200 | — |
| 42 | ▼ 6 | qwen2-72b-instruct开源Alibaba | 1,240 | ±2 | 38,873 | — |
| 43 | ▼ 6 | phi-4开源Microsoft | 1,237 | ±3 | 25,213 | — |
| 44 | ▼ 6 | command-r-08-2024开源Cohere | 1,236 | ±5 | 10,852 | — |
| 45 | ▼ 6 | olmo-2-0325-32b-instruct开源Ai2 | 1,236 | ±10 | 3,456 | — |
| 46 | ▼ 6 | jamba-1.5-mini开源AI21 Labs | 1,220 | ±6 | 9,273 | — |
| 47 | ▼ 6 | ministral-8b-2410开源Mistral | 1,220 | ±8 | 5,109 | — |
| 48 | ▼ 6 | qwen1.5-110b-chat开源Alibaba | 1,215 | ±3 | 27,431 | — |
| 49 | ▼ 6 | c4ai-aya-expanse-8b开源Cohere | 1,211 | ±5 | 10,391 | — |
| 50 | ▼ 6 | mixtral-8x22b-instruct-v0.1开源Mistral | 1,209 | ±3 | 53,756 | — |
智力进步
涨得最多 · Anthropic +57.2目前最高 · Anthropic 64.6最高与最低相差 18.1 分时间范围 2025年11月 – 2026年11月
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi