模型排行
130 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1,251
llama-3.1-405b-instruct-bf16Meta · 总第 8 名1,251
mistral-large-2407Mistral · 总第 19 名1,229
| 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 | 1,251 | ±3 | 43,099 | — |
| 2 | — | llama-3.1-405b-instruct-bf16开源Meta | 1,251 | ±5 | 14,607 | — |
| 3 | — | mistral-large-2407开源Mistral | 1,229 | ±3 | 39,456 | — |
| 4 | — | qwen-max-0919开源Alibaba | 1,228 | ±5 | 15,384 | — |
| 5 | ▲ 1 | deepseek-v2.5开源DeepSeek | 1,221 | ±4 | 17,601 | — |
| 6 | ▲ 1 | athene-70b-0725开源NexusFlow | 1,221 | ±4 | 20,711 | — |
| 7 | ▼ 2 | qwen2.5-72b-instruct开源Alibaba | 1,220 | ±4 | 17,390 | — |
| 8 | ▲ 1 | llama-3.1-nemotron-70b-instruct开源NVIDIA | 1,215 | ±9 | 6,228 | — |
| 9 | 新 | llama-3.1-nemotron-51b-instruct开源NVIDIA | 1,212 | ±13 | 1,816 | — |
| 10 | ▼ 2 | llama-3.1-70b-instruct开源Meta | 1,211 | ±3 | 38,692 | — |
| 11 | ▼ 1 | gemma-2-27b-it开源Google | 1,205 | ±3 | 49,844 | — |
| 12 | ▼ 1 | jamba-1.5-large开源AI21 Labs | 1,203 | ±6 | 9,182 | — |
| 13 | ▼ 1 | gemma-2-9b-it-simpo开源Princeton | 1,197 | ±6 | 10,622 | — |
| 14 | ▼ 1 | command-r-plus-08-2024开源Cohere | 1,196 | ±6 | 10,599 | — |
| 15 | ▼ 1 | nemotron-4-340b-instruct开源NVIDIA | 1,196 | ±5 | 20,649 | — |
| 16 | ▼ 1 | llama-3-70b-instruct开源Meta | 1,194 | ±2 | 163,932 | — |
| 17 | ▼ 1 | command-r-plus开源Cohere | 1,183 | ±3 | 80,899 | — |
| 18 | ▼ 1 | deepseek-coder-v2开源DeepSeek | 1,181 | ±5 | 15,804 | — |
| 19 | ▼ 1 | qwen2-72b-instruct开源Alibaba | 1,180 | ±3 | 38,990 | — |
| 20 | ▼ 1 | gemma-2-9b-it开源Google | 1,176 | ±3 | 36,551 | — |
| 21 | ▼ 1 | command-r-08-2024开源Cohere | 1,175 | ±6 | 10,940 | — |
| 22 | ▼ 1 | jamba-1.5-mini开源AI21 Labs | 1,158 | ±6 | 9,299 | — |
| 23 | ▼ 1 | qwen1.5-110b-chat开源Alibaba | 1,155 | ±4 | 27,486 | — |
| 24 | ▼ 1 | mixtral-8x22b-instruct-v0.1开源Mistral | 1,149 | ±3 | 53,840 | — |
| 25 | ▼ 1 | command-r开源Cohere | 1,148 | ±3 | 56,405 | — |
| 26 | ▼ 1 | qwen1.5-72b-chat开源Alibaba | 1,148 | ±3 | 40,647 | — |
| 27 | ▼ 1 | llama-3-8b-instruct开源Meta | 1,142 | ±3 | 109,378 | — |
| 28 | ▼ 1 | yi-1.5-34b-chat开源01 AI | 1,132 | ±4 | 25,145 | — |
| 29 | ▼ 1 | zephyr-orpo-141b-A35b-v0.1开源HuggingFace | 1,131 | ±7 | 4,857 | — |
| 30 | ▼ 1 | llama-3.1-8b-instruct开源Meta | 1,127 | ±3 | 36,916 | — |
| 31 | ▼ 1 | qwen1.5-32b-chat开源Alibaba | 1,127 | ±4 | 22,763 | — |
| 32 | ▼ 1 | phi-3-medium-4k-instruct开源Microsoft | 1,118 | ±4 | 26,158 | — |
| 33 | ▼ 1 | internlm2_5-20b-chat开源InternLM | 1,115 | ±5 | 10,710 | — |
| 34 | ▼ 1 | mixtral-8x7b-instruct-v0.1开源Mistral | 1,114 | ±0 | 76,142 | — |
| 35 | ▼ 1 | dbrx-instruct-preview开源Databricks | 1,113 | ±3 | 33,730 | — |
| 36 | ▼ 1 | gemma-2-2b-it开源Google | 1,110 | ±4 | 28,101 | — |
| 37 | ▼ 1 | qwen1.5-14b-chat开源Alibaba | 1,107 | ±4 | 18,674 | — |
| 38 | ▼ 1 | wizardlm-70b开源Microsoft | 1,104 | ±7 | 8,384 | — |
| 39 | ▼ 1 | snowflake-arctic-instruct开源Snowflake | 1,103 | ±3 | 34,178 | — |
| 40 | ▼ 1 | yi-34b-chat开源01 AI | 1,101 | ±5 | 15,928 | — |
| 41 | ▼ 1 | openchat-3.5-0106开源OpenChat · 另有 1 个档位 | 1,100 | ±6 | 12,975 | — |
| 42 | ▼ 1 | phi-3-small-8k-instruct开源Microsoft | 1,098 | ±5 | 18,500 | — |
| 43 | ▼ 1 | gemma-1.1-7b-it开源Google | 1,098 | ±4 | 25,082 | — |
| 44 | ▼ 1 | llama-3.2-3b-instruct开源Meta | 1,093 | ±7 | 8,459 | — |
| 45 | ▼ 1 | tulu-2-dpo-70b开源AllenAI/UW | 1,092 | ±7 | 6,668 | — |
| 46 | ▼ 1 | vicuna-33b开源LMSYS | 1,091 | ±4 | 22,942 | — |
| 47 | ▼ 1 | deepseek-llm-67b-chat开源DeepSeek | 1,091 | ±8 | 4,984 | — |
| 48 | ▼ 1 | starling-lm-7b-beta开源Nexusflow | 1,091 | ±4 | 16,673 | — |
| 49 | ▼ 1 | openhermes-2.5-mistral-7b开源NousResearch | 1,090 | ±8 | 5,092 | — |
| 50 | ▼ 1 | starling-lm-7b-alpha开源UC Berkeley | 1,083 | ±5 | 10,416 | — |
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