Rank,LLM,Organization,EI Score,Release date (MM/YY),Chatbot Arena Elo,"MMLU
(5-shot)",Size,License
2,gpt-4o-2024-05-13,Open AI,"59.2 (-1.2, 1.7)",05/24,1287,88.7,🔒,Proprietary
3,gpt-4-turbo-04-09,Open AI,"57.5 (-1.2, 1.7)",04/24,1256,🔒,🔒,Proprietary
18,gpt-4-0613,Open AI,"26.9 (-1.4, 1.4)",06/23,1246,86.4,🔒,Proprietary
19,gpt-3.5-turbo-0125,Open AI,"18.2 (-1.1, 1.1)",01/24,1102,70.0,🔒,Proprietary
15,mistral-large-latest,Mistral,"33.9 (-1.5, 1.3)",02/24,1156,81.2,🔒,Proprietary
17,open-mixtral-8x22b,Mistral,"29.3 (-1.6, 1.5)",04/24,1146,77.8,176 B,Apache 2.0
14,open-mixtral-8x7b,Mistral,"34.4 (-1.4, 1.5)",12/23,1114,70.6,56 B,Apache 2.0
8,llama-3-70b-chat-hf,Meta,"45.1 (-1.5, 1.4)",04/24,1208,82.0,70 B,Llama 3 Community
16,llama-3-8b-chat-hf,Meta,"30.0 (-1.4, 1.4)",04/24,1153,68.4,8 B,Llama 3 Community
20,gemini-1.5-pro-preview-0409,Google,"11.6 (-0.9, 0.8)",05/24,1268,81.9,🔒,Proprietary
4,gemini-1.0-pro,Google,"50.6 (-1.2, 1.5)",04/24,1208,71.8,🔒,Proprietary
10,dbrx,Databricks,"38.8 (-1.5, 1.9)",03/24,1103,73.7,132 B,DBRX LICENSE
12,command-r-plus,Cohere,"35.6 (-1.7, 1.7)",04/24,1189,75.7,104 B,CC-BY-NC-4.0
13,command-r,Cohere,"34.7 (-1.7, 1.5)",04/24,1147,68.2,35 B,CC-BY-NC-4.0
5,claude-3-opus-20240229,Anthropic,"50.1 (-1.5, 1.4)",03/24,1248,86.8,🔒,Proprietary
9,claude-3-sonnet-20240229,Anthropic,"42.5 (-1.5, 1.6)",03/24,1201,79.0,🔒,Proprietary
11,claude-3-haiku-20240307,Anthropic,"38.6 (-1.7, 2.2)",03/24,1178,75.2,🔒,Proprietary
1,qwen1.5-110B-chat,Alibaba,"65.6 (-1.2, 1.8)",02/24,1164,80.2,100 B,Qianwen LICENSE
7,qwen1.5-72B-chat,Alibaba,"48.7 (-1.4, 1.6)",02/24,1152,77.4,72 B,Qianwen LICENSE
6,qwen1.5-32B-chat,Alibaba,"50.0 (0.0, 0.0)",02/24,1126,74.3,32 B,Qianwen LICENSE