LM Benchmarks Logo

LM Benchmarks

Beta
  Model  Size  Release  Updated  LBCode  Aider  HumEval  HumEval+  ArenaCode  LBReason  MMLU Pro  MMLU  GPQA  MATH  LBMath  Arena  AA Score  In Cost  Out Cost  Cutoff  Context  Org  Type 
OpenAI logo
o1Unknown69.7%84.2%91.6%76.0%67.3%OpenAIProprietary
Anthropic logo
Claude 3.5 SonnetUnknown67.1%84.2%93.7%85.9%132158.7%78.0%88.0%65.0%78.3%51.3%128280$3$15Apr 24200KAnthropicProprietary
Google logo
Gemini Exp 1206Unknown63.4%80.5%138557.0%70.0%13792MGoogleProprietary
Anthropic logo
Claude 3.5 SonnetUnknown60.9%77.4%92.0%129558.7%76.1%88.7%59.4%71.1%53.3%126876$3$15Apr 24200KAnthropicProprietary
Alibaba logo
Qwen 2.572BUnknown57.6%55.6%86.6%51.2%128346.0%71.1%86.8%49.0%83.1%52.4%125975$0.38$0.57Sep 24128KAlibabaOpen Source
Alibaba logo
Qwen 2.5 Coder32BUnknown56.9%72.9%92.7%86.3%126447.3%79.0%41.0%72.0%46.0%122070$0.2$0.2Sep 24128KAlibabaOpen Source
Google logo
Gemini 2.0 Flash ExpUnknown54.4%69.9%91.0%134159.1%76.4%87.0%62.1%89.7%59.4%135682Aug 241MGoogleProprietary
Google logo
Gemini 2 Flash ThinkingUnknown53.1%134364.6%69.0%136932KGoogleProprietary
Google logo
Gemini Exp 1114Unknown52.4%60.9%132454.7%54.9%134332KGoogleProprietary
OpenAI logo
GPT 4oUnknown51.4%71.4%90.0%87.2%127454.7%74.7%89.0%51.0%78.0%48.2%126577$2.5$10Oct 23128KOpenAIProprietary
Anthropic logo
Claude 3.5 HaikuUnknown51.4%75.2%88.1%126929.3%65.0%81.0%41.6%73.0%35.5%123969$1$5Apr 24200KAnthropicProprietary
OpenAI logo
o1 PreviewUnknown50.8%79.7%92.4%89.0%135568.0%91.0%73.3%85.0%62.9%133385$15$60Oct 23128KOpenAIProprietary
DeepSeek logo
Deepseek 3671BUnknown79.7%82.6%75.9%88.5%59.1%$0.07$1.1128KDeepSeekOpen Source
Google logo
Gemini Exp 1121Unknown50.4%57.9%134145.3%62.7%136532KGoogleProprietary
Google logo
Gemini 1.5 Pro 002Unknown48.8%66.9%84.1%79.3%129046.0%75.8%86.0%59.1%86.5%57.4%130180$1.25$2.5Nov 232MGoogleProprietary
OpenAI logo
o1 miniUnknown48.0%70.7%92.4%89.0%136277.3%85.0%60.0%90.0%59.2%130882$3$12Oct 23128KOpenAIProprietary
Mistral logo
Mistral Large123BUnknown47.1%65.4%90.0%127743.4%84.0%47.0%72.0%32.6%125174$2$6Jul 24128KMistralProprietary
Mistral logo
Mistral Large123BUnknown47.1%60.2%91.3%62.2%126942.0%65.9%85.0%48.0%72.0%43.7%125074$2$6Jul 24128KMistralProprietary
DeepSeek logo
DeepSeek 2.5236BUnknown46.3%72.2%89.0%83.5%128839.3%65.8%81.0%42.0%55.0%122166$0.14$0.28Nov 23128KDeepSeekProprietary
OpenAI logo
GPT 4oUnknown46.1%72.9%90.0%135153.3%74.7%86.0%39.0%69.0%42.5%136171$2.5$10Oct 23128KOpenAIProprietary
xAI logo
Grok 2Unknown45.1%58.6%88.4%80.5%128736.3%75.5%87.5%56.0%76.1%45.8%129070$5$15Aug 24128KxAIProprietary
OpenAI logo
GPT 4o miniUnknown43.1%55.6%87.2%83.5%128435.3%63.1%82.0%43.0%75.0%35.6%127271$0.15$0.6Oct 23128KOpenAIProprietary
Meta logo
Llama 3.1405BUnknown42.6%66.2%89.0%128453.3%73.3%87.0%51.1%73.8%40.5%126672$5$15Dec 23128KMetaOpen Source
Google logo
Gemini 1.5 Flash 002Unknown41.9%52.6%84.0%75.6%125450.0%67.3%81.0%51.0%77.9%47.2%126973$0.075$0.3Nov 231MGoogleProprietary
DeepSeek logo
DeepSeek Coder 2236BUnknown41.5%72.9%90.2%82.3%126745.3%63.6%80.0%42.0%58.0%121367$0.14$0.28Nov 23128KDeepSeekProprietary
Google logo
Gemini 1.5 Pro 001Unknown40.9%66.9%126249.3%69.0%56.1%1260Nov 232MGoogleProprietary
Google logo
Gemini 1.5 Flash 001Unknown40.6%52.6%123247.3%59.1%1227Nov 231MGoogleProprietary
Anthropic logo
Claude 3 OpusUnknown38.6%68.4%84.9%77.4%125041.3%68.5%84.0%50.4%60.1%43.4%124870$15$75Aug 23200KAnthropicProprietary
Alibaba logo
Qwen 2.57BUnknown38.4%63.9%26.2%38.2%Sep 24128KAlibabaOpen Source
xAI logo
Grok-2 miniUnknown37.5%54.9%85.7%126242.0%72.0%86.2%51.0%73.0%1266Aug 24128KxAIProprietary
Alibaba logo
QwQ Preview32BUnknown37.2%57.7%75.0%65.2%50.8%Sep 2432KAlibabaOpen Source
Amazon logo
Nova ProUnknown37.1%54.1%89.0%32.6%84.0%46.9%76.6%39.1%75$0.8$3.2300KAmazonProprietary
Meta logo
Llama 3.370BUnknown36.6%59.4%88.4%87.6%138550.7%68.9%86.0%50.5%77.0%41.1%74$0.1$0.4MetaOpen Source
Google logo
Gemma 227BUnknown35.9%51.8%28.1%56.5%75.2%42.3%26.2%Jun 248KGoogleOpen Source
Meta logo
Llama 3.170BUnknown32.7%58.6%80.5%124440.7%66.4%86.0%43.0%60.0%34.4%124265$1$3Dec 23128KMetaOpen Source
Google logo
Gemini 1.5 Flash8BUnknown28.7%38.3%116333.3%58.7%38.4%58.7%1177$0.0375$0.15Nov 231MGoogleProprietary
Mistral logo
Mistral Nemo12BUnknown28.7%33.1%25.3%44.8%68.0%33.0%40.0%16.9%52$0.15$0.15Apr 24128KMistralProprietary
Amazon logo
Nova LiteUnknown27.5%85.4%34.7%79.0%42.0%73.3%30.8%70$0.06$0.24300KAmazonProprietary
Alibaba logo
Qwen 2.532BUnknown27.0%58.6%88.4%69.2%83.9%49.5%83.1%Sep 24128KAlibabaOpen Source
Anthropic logo
Claude 3 HaikuUnknown24.5%47.4%75.9%118925.9%42.3%75.2%33.3%40.9%22.9%117954$0.25$1.25Aug 23200KAnthropicProprietary
Google logo
Gemma 29BUnknown22.5%40.2%15.2%52.1%71.3%36.6%19.5%Jun 248KGoogleOpen Source
Alibaba logo
Qwen 2.514BUnknown20.9%83.5%63.7%80.0%45.5%80.0%Sep 24128KAlibabaOpen Source
Alibaba logo
Qwen 2.5 Coder14BUnknown23.4%69.2%89.6%84.0%Sep 24128KAlibabaOpen Source
Alibaba logo
Qwen 2.5 Coder7BUnknown20.3%63.9%88.4%81.9%Sep 24128KAlibabaOpen Source
Amazon logo
Nova MicroUnknown20.2%81.1%25.1%76.0%40.0%69.3%33.7%66$0.04$0.14128KAmazonProprietary
Alibaba logo
Qwen 2.5 Coder3BUnknown10.8%39.1%84.1%75.2%Sep 2432KAlibabaOpen Source
Meta logo
Llama 3.18BUnknown19.7%37.6%72.6%62.8%117415.3%44.3%71.0%27.0%64.0%16.6%117153$0.3$0.6Dec 23128KMetaOpen Source
Alibaba logo
Qwen 2.5 Coder1.5BUnknown6.1%31.6%70.7%66.5%Sep 2432KAlibabaOpen Source
Alibaba logo
Qwen 2.53BUnknown43.7%Sep 2432KAlibabaOpen Source
Alibaba logo
Qwen 2.51.5BUnknown4.7%32.1%Sep 2432KAlibabaOpen Source
Alibaba logo
Qwen 2.50.5BUnknown2.0%14.9%Sep 2432KAlibabaOpen Source
Alibaba logo
Qwen 2.5 Coder0.5BUnknown2.0%14.3%61.6%53.8%Sep 2432KAlibabaOpen Source
Mistral logo
Pixtral Large123BUnknown82.0%85.0%52.0%72.0%73$2$6128KMistralProprietary
Meta logo
Llama 3.23BUnknown26.3%104822.2%63.4%21.0%49.0%109346$0.08$0.1Dec 23128KMetaOpen Source
Meta logo
Llama 3.21BUnknown100911.9%49.3%14.0%27.0%104428$0.04$0.08Dec 23128KMetaOpen Source
Meta logo
Llama 3.2 (Vision)90BUnknown86.0%43.0%68.0%67$0.9$0.9Dec 23128KMetaOpen Source
Meta logo
Llama 3.2 (Vision)11BUnknown73.0%26.0%51.9%54$0.19$0.19Dec 23128KMetaOpen Source
DeepSeek logo
R1 Lite PreviewUnknown58.5%91.6%DeepSeekProprietary
OpenAI logo
o1 Pro modeUnknown79.0%OpenAIProprietary
Google logo
Gemma 22BUnknown17.7%51.3%15.0%Jun 248KGoogleOpen Source
Alibaba logo
Qwen 2.5 TurboUnknownSep 241MAlibabaProprietary
DeepSeek logo
Deepseek 2.5UnknownNov 23128KDeepSeekProprietary
Phi 414BUnknown82.6%82.8%70.4%84.8%56.1%80.4%MicrosoftOpen Source
OpenAI logo
o3Unknown87.7%OpenAIProprietary
OpenAI logo
o3 mini highUnknownOpenAIProprietary
OpenAI logo
o3 mini mediumUnknownOpenAIProprietary
OpenAI logo
o3 mini lowUnknownOpenAIProprietary
67 models