diff --git a/README.md b/README.md index b46c970..02b464e 100644 --- a/README.md +++ b/README.md @@ -18,6 +18,6 @@ These are shown in green 🟢 and are the best LLMs to use. Some LLMs are "pareto suboptimal", i.e. there is no LLM worse in both cost and quality. These are shown in red 🔴 and are the LLMs to avoid. -**Last updated: 24 July 2024** +**Last updated: 4 Aug 2024** Alternatives: [ArtificialAnalysis.ai](https://artificialanalysis.ai/) diff --git a/elo.csv b/elo.csv index 6b249f2..a7d0dba 100644 --- a/elo.csv +++ b/elo.csv @@ -1,16 +1,20 @@ model,elo,cpmi,source -gpt-4o-2024-05-13,1287,5,https://openai.com/api/pricing/ +gpt-4o-2024-05-13,1286,5,https://openai.com/api/pricing/ gpt-4o-mini-2024-07-18,1280,0.15,https://openai.com/api/pricing/ claude-3-5-sonnet-20240620,1271,3, -gemini-advanced-0514,1267,, -gemini-1.5-pro-api-0514,1263,3.5,https://ai.google.dev/pricing -gemini-1.5-pro-api-0409-preview,1258,7,https://llmpricecheck.com/google/gemini-1.5-pro -gpt-4-turbo-2024-04-09,1256,10,https://openai.com/api/pricing/ -gpt-4-1106-preview,1252,10,https://openai.com/api/pricing/ +gemini-advanced-0514,1266,, +meta-llama-3.1-405b-instruct,1262,2,https://openrouter.ai/models/meta-llama/llama-3.1-405b +gemini-1.5-pro-api-0514,1261,3.5,https://ai.google.dev/pricing +gemini-1.5-pro-api-0409-preview,1257,7,https://llmpricecheck.com/google/gemini-1.5-pro +gpt-4-turbo-2024-04-09,1257,10,https://openai.com/api/pricing/ +gpt-4-1106-preview,1251,10,https://openai.com/api/pricing/ claude-3-opus-20240229,1248,15, -gpt-4-0125-preview,1246,10,https://openai.com/api/pricing/ +athene-70b,1245,, +meta-llama-3.1-70b-instruct,1242,0.52,https://openrouter.ai/models/meta-llama/llama-3.1-70b-instruct +gpt-4-0125-preview,1245,10,https://openai.com/api/pricing/ yi-large-preview,1239,3,https://openrouter.ai/models/01-ai/yi-large -gemini-1.5-flash-api-0514,1232,0.35,https://ai.google.dev/pricing +gemini-1.5-flash-api-0514,1227,0.35,https://ai.google.dev/pricing +deepseek-v2-api-0628,1221,0.14, gemma-2-27b-it,1212,0.27,https://openrouter.ai/models/google/gemma-2-27b-it yi-large,1239,3,https://openrouter.ai/models/01-ai/yi-large bard-jan-24-gemini-pro,1208,7,https://ai.google.dev/pricing @@ -25,11 +29,12 @@ gpt-4-0314,1186,30,https://platform.openai.com/docs/deprecations/ gemma-2-9b-it,1186,0.2,https://openrouter.ai/models/google/gemma-2-9b-it glm-4-0116,1183,, qwen-max-0428,1183,, -deepseek-coder-v2-instruct,1179,0.14,https://openrouter.ai/models/deepseek/deepseek-coder +deepseek-coder-v2-instruct,1178,0.14,https://openrouter.ai/models/deepseek/deepseek-coder claude-3-haiku-20240307,1178,0.25, +meta-llama-3.1-8b-instruct,1166,0.06,https://openrouter.ai/models/meta-llama/llama-3.1-8b-instruct qwen1.5-110b-chat,1164,1.62,https://openrouter.ai/models/qwen/qwen-110b-chat gpt-4-0613,1161,30,https://platform.openai.com/docs/deprecations/ -yi-1.5-34b-chat,1158, +yi-1.5-34b-chat,1158,, mistral-large-2402,1156,8, llama-3-8b-instruct,1153,0.05, claude-1,1149,8,