模型排行
147 个模型 · 1 个榜单 · 1 个数据来源
人气王座
开源之光
国产之光
最多人测
开源前三甲
llama-3.1-405b-instruct-fp8Meta · 总第 7 名1,252
llama-3.1-405b-instruct-bf16Meta · 总第 8 名1,252
llama-3.3-70b-instructMeta · 总第 18 名1,234
| 排名 | 变化 | 模型 | 得分 | 置信区间 | 投票 | 价格 $/M |
|---|---|---|---|---|---|---|
| 1 | — | llama-3.1-405b-instruct-fp8开源Meta | 1,252 | ±3 | 61,518 | — |
| 2 | — | llama-3.1-405b-instruct-bf16开源Meta | 1,252 | ±5 | 14,535 | — |
| 3 | 新 | llama-3.3-70b-instruct开源Meta | 1,234 | ±7 | 7,647 | — |
| 4 | ▼ 1 | athene-v2-chat开源NexusFlow | 1,232 | ±4 | 12,597 | — |
| 5 | ▼ 1 | mistral-large-2407开源Mistral · 另有 1 个档位 | 1,231 | ±3 | 48,378 | — |
| 6 | ▼ 1 | qwen-max-0919开源Alibaba | 1,231 | ±4 | 17,581 | — |
| 7 | ▼ 1 | deepseek-v2.5开源DeepSeek | 1,225 | ±3 | 26,507 | — |
| 8 | — | athene-70b-0725开源NexusFlow | 1,223 | ±3 | 20,643 | — |
| 9 | ▼ 2 | qwen2.5-72b-instruct开源Alibaba | 1,222 | ±4 | 35,637 | — |
| 10 | ▼ 1 | llama-3.1-nemotron-70b-instruct开源NVIDIA | 1,216 | ±5 | 7,672 | — |
| 11 | ▼ 1 | llama-3.1-70b-instruct开源Meta | 1,214 | ±3 | 57,780 | — |
| 12 | ▼ 1 | gemma-2-27b-it开源Google | 1,207 | ±3 | 67,092 | — |
| 13 | ▼ 1 | jamba-1.5-large开源AI21 Labs | 1,204 | ±6 | 9,145 | — |
| 14 | ▼ 1 | llama-3.1-nemotron-51b-instruct开源NVIDIA | 1,204 | ±8 | 3,928 | — |
| 15 | ▼ 1 | gemma-2-9b-it-simpo开源Princeton | 1,199 | ±6 | 10,569 | — |
| 16 | ▼ 1 | command-r-plus-08-2024开源Cohere | 1,197 | ±5 | 10,578 | — |
| 17 | ▼ 1 | nemotron-4-340b-instruct开源NVIDIA | 1,196 | ±4 | 20,648 | — |
| 18 | ▼ 1 | llama-3-70b-instruct开源Meta | 1,195 | ±2 | 163,884 | — |
| 19 | ▼ 1 | qwen2.5-coder-32b-instruct开源Alibaba | 1,189 | ±6 | 5,744 | — |
| 20 | ▼ 1 | c4ai-aya-expanse-32b开源Cohere | 1,183 | ±4 | 22,447 | — |
| 21 | ▼ 1 | command-r-plus开源Cohere | 1,183 | ±2 | 80,936 | — |
| 22 | ▼ 1 | deepseek-coder-v2开源DeepSeek | 1,182 | ±5 | 15,785 | — |
| 23 | ▼ 1 | qwen2-72b-instruct开源Alibaba | 1,181 | ±4 | 38,962 | — |
| 24 | ▼ 1 | gemma-2-9b-it开源Google | 1,179 | ±3 | 45,891 | — |
| 25 | ▼ 1 | command-r-08-2024开源Cohere | 1,176 | ±5 | 10,876 | — |
| 26 | — | jamba-1.5-mini开源AI21 Labs | 1,159 | ±6 | 9,282 | — |
| 27 | ▼ 2 | ministral-8b-2410开源Mistral | 1,159 | ±8 | 5,142 | — |
| 28 | ▼ 1 | qwen1.5-110b-chat开源Alibaba | 1,156 | ±4 | 27,469 | — |
| 29 | 新 | c4ai-aya-expanse-8b开源Cohere | 1,153 | ±8 | 3,291 | — |
| 30 | ▼ 2 | mixtral-8x22b-instruct-v0.1开源Mistral | 1,149 | ±3 | 53,828 | — |
| 31 | ▼ 2 | qwen1.5-72b-chat开源Alibaba | 1,148 | ±3 | 40,649 | — |
| 32 | ▼ 2 | command-r开源Cohere | 1,148 | ±3 | 56,401 | — |
| 33 | ▼ 2 | llama-3-8b-instruct开源Meta | 1,143 | ±2 | 109,298 | — |
| 34 | ▼ 2 | yi-1.5-34b-chat开源01 AI | 1,133 | ±4 | 25,157 | — |
| 35 | ▼ 2 | llama-3.1-8b-instruct开源Meta | 1,133 | ±3 | 51,800 | — |
| 36 | ▼ 2 | zephyr-orpo-141b-A35b-v0.1开源HuggingFace | 1,131 | ±8 | 4,860 | — |
| 37 | ▼ 2 | qwen1.5-32b-chat开源Alibaba | 1,127 | ±4 | 22,761 | — |
| 38 | ▼ 2 | phi-3-medium-4k-instruct开源Microsoft | 1,118 | ±4 | 26,130 | — |
| 39 | ▼ 2 | internlm2_5-20b-chat开源InternLM | 1,116 | ±6 | 10,676 | — |
| 40 | ▼ 1 | gemma-2-2b-it开源Google | 1,115 | ±4 | 37,482 | — |
| 41 | ▼ 3 | mixtral-8x7b-instruct-v0.1开源Mistral | 1,114 | ±0 | 76,146 | — |
| 42 | ▼ 2 | dbrx-instruct-preview开源Databricks | 1,113 | ±3 | 33,745 | — |
| 43 | ▼ 2 | granite-3.0-8b-instruct开源IBM | 1,110 | ±8 | 7,068 | — |
| 44 | ▼ 2 | qwen1.5-14b-chat开源Alibaba | 1,107 | ±5 | 18,682 | — |
| 45 | ▼ 2 | wizardlm-70b开源Microsoft | 1,105 | ±7 | 8,383 | — |
| 46 | ▼ 2 | snowflake-arctic-instruct开源Snowflake | 1,102 | ±3 | 34,177 | — |
| 47 | ▼ 2 | yi-34b-chat开源01 AI | 1,101 | ±4 | 15,923 | — |
| 48 | ▼ 2 | openchat-3.5-0106开源OpenChat · 另有 1 个档位 | 1,100 | ±5 | 12,987 | — |
| 49 | ▼ 2 | gemma-1.1-7b-it开源Google | 1,099 | ±3 | 25,080 | — |
| 50 | ▼ 2 | phi-3-small-8k-instruct开源Microsoft | 1,098 | ±4 | 18,501 | — |
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