LogoTop AI Hubs

LLM Leaderboard - Comparison of AI Models

Comparison and ranking the performance of over 30 AI models (LLMs) across key metrics including quality, price, performance and speed (output speed - tokens per second & latency - TTFT), context window & others. For more details including relating to our methodology, see our FAQs.

ModelProviderLicenseAI IndexMMLU-ProGPQALiveCodeContextPrice/1MTokens/sTTFT (s)
o4-mini (high)OpenAIProprietary7083%78%80%200k$1.93150.332.62
Gemini 2.5 ProGoogleProprietary6886%84%70%1m$3.44156.936.54
o3OpenAIProprietary6785%83%53%128k$17.50197.713
Grok 3 mini Reasoning (high)xAIProprietary6783%79%70%1m$0.35115.617.52
o3-mini (high)OpenAIProprietary6680%77%73%200k$1.93151.449.75
o3-miniOpenAIProprietary6379%75%72%200k$1.93138.814.22
Qwen3 235B A22B (Reasoning)AlibabaOpen6283%70%62%128k$0.3039.351.51
o1OpenAIProprietary6284%75%68%200k$26.2511525.5
Llama 3.1 Nemotron Ultra 253B ReasoningNVIDIAOpen6183%73%64%128k$0.00
Gemini 2.5 Flash (Reasoning)GoogleProprietary6080%70%51%1m$0.99342.98.15
DeepSeek R1DeepSeekOpen6084%71%62%128k$0.9623.8102.63
Qwen3 32B (Reasoning)AlibabaOpen5980%67%55%128k$0.1734.558.59
QwQ-32BAlibabaOpen5876%59%63%131k$0.4789.728.19
Claude 3.7 Sonnet ThinkingAnthropicProprietary5784%77%47%200k$6.00
Qwen3 14B (Reasoning)AlibabaOpen5677%60%52%128k$0.1285.224
Qwen3 30B A3B (Reasoning)AlibabaOpen5678%62%51%128k$0.19112.818.26
o1-miniOpenAIProprietary5474%60%58%128k$1.93206.59.95
DeepSeek V3 (Mar' 25)DeepSeekOpen5382%66%41%128k$0.4826.23.42
GPT-4.1 miniOpenAIProprietary5378%66%48%1m$0.7075.80.56
GPT-4.1OpenAIProprietary5381%67%46%1m$3.50136.50.57
Gemini 2.0 Flash Thinking exp. (Jan '25)GoogleProprietary5280%70%32%1m$0.00
DeepSeek R1 Distill Qwen 32BDeepSeekOpen5274%62%27%128k$0.2243.446.61
Llama 3.3 Nemotron Super 49B ReasoningNVIDIAOpen5179%64%28%128k$0.00
Grok 3xAIProprietary5180%69%43%1m$6.0050.70.49
Llama 4 MaverickMetaOpen5181%67%40%1m$0.35129.30.34
GPT-4o (March 2025)OpenAIProprietary5080%66%43%128k$7.50142.60.32
Gemini 2.0 Pro ExperimentalGoogleProprietary4981%62%35%2m$0.0025.216.62
DeepSeek R1 Distill Qwen 14BDeepSeekOpen4974%48%38%128k$0.88128.116.06
Gemini 2.5 FlashGoogleProprietary4878%59%37%1m$0.26265.40.28
DeepSeek R1 Distill Llama 70BDeepSeekOpen4880%40%27%128k$0.60115.917.66
Claude 3.7 SonnetAnthropicProprietary4880%66%39%200k$6.00781.03
Gemini 2.0 FlashGoogleProprietary4878%62%33%1m$0.17232.60.37
Qwen3 4B (Reasoning)AlibabaOpen4770%52%47%32k$0.00
Reka Flash 3Reka AIOpen4767%53%44%128k$0.3556.736.24
Gemini 2.0 Flash (exp)GoogleProprietary4678%64%21%1m$0.00236.50.25
DeepSeek V3 (Dec '24)DeepSeekOpen4675%56%36%128k$0.4826.63.52
Qwen2.5 MaxAlibabaProprietary4576%59%36%32k$2.80501.24
Gemini 1.5 Pro (Sep)GoogleProprietary4575%59%32%2m$2.1992.80.65
Claude 3.5 Sonnet (Oct)AnthropicProprietary4477%60%38%200k$6.0078.61.12
SonarPerplexityProprietary4369%47%30%127k$1.00148.41.67
Llama 4 ScoutMetaOpen4375%59%30%10m$0.27120.10.36
Sonar ProPerplexityProprietary4376%58%28%200k$6.0087.52.27
QwQ 32B-PreviewAlibabaOpen4365%56%34%33k$0.2656.236.06
Nova PremierAmazonProprietary4373%57%32%1m$5.0067.20.87
GPT-4o (Nov '24)OpenAIProprietary4175%54%31%128k$4.38137.20.55
Gemini 2.0 Flash-Lite (Feb '25)GoogleProprietary4172%54%19%1m$0.13209.50.27
Llama 3.3 70BMetaOpen4171%50%29%128k$0.60126.70.44
GPT-4.1 nanoOpenAIProprietary4166%51%33%1m$0.17198.50.44
GPT-4o (May '24)OpenAIProprietary4174%53%33%128k$7.5081.90.57
Llama 3.1 405BMetaOpen4073%52%31%128k$3.5032.30.67
Qwen2.5 72BAlibabaOpen4072%49%28%131k$0.0041.21.18
MiniMax-Text-01MiniMaxOpen4076%58%25%4m$0.4231.90.82
Phi-4Microsoft AzureOpen4071%57%23%16k$0.2239.90.44
Command ACohereOpen4071%53%29%256k$4.38100.60.21
Tulu3 405BAllen Institute for AIOpen4072%52%29%128k$0.00
Llama 3.3 Nemotron Super 49B v1NVIDIAOpen3970%52%28%128k$0.00
Grok 2xAIProprietary3971%51%27%131k$0.00
Gemini 1.5 Flash (Sep)GoogleProprietary3968%46%27%1m$0.13194.20.2
Mistral Large 2 (Nov '24)MistralOpen3870%49%29%128k$3.0073.20.4
Qwen3 1.7B (Reasoning)AlibabaOpen3857%36%31%32k$0.00
Gemma 3 27BGoogleOpen3867%43%14%128k$0.00
Grok BetaxAIProprietary3870%47%24%128k$7.5067.50.3
Pixtral LargeMistralOpen3770%51%26%128k$3.0041.30.36
Qwen2.5 Instruct 32BAlibabaOpen3770%47%25%128k$0.15
Llama 3.1 Nemotron 70BNVIDIAOpen3769%47%17%128k$0.2443.20.55
Nova ProAmazonProprietary3769%50%23%300k$1.40
Mistral Large 2 (Jul '24)MistralOpen3768%47%27%128k$3.0036.20.47
Qwen2.5 Coder 32BAlibabaOpen3664%42%30%131k$0.2052.60.52
GPT-4o miniOpenAIProprietary3665%43%23%128k$0.2668.10.36
Llama 3.1 70BMetaOpen3568%41%23%128k$0.7265.60.51
Mistral Small 3.1MistralOpen3566%45%21%128k$0.15167.80.27
Mistral Small 3MistralOpen3565%46%25%32k$0.15130.80.3
Claude 3 OpusAnthropicProprietary3570%49%28%200k$30.0026.60.99
Claude 3.5 HaikuAnthropicProprietary3563%41%31%200k$1.6066.20.83
DeepSeek R1 Distill Llama 8BDeepSeekOpen3454%30%23%128k$0.0451.639.48
Gemma 3 12BGoogleOpen3460%35%14%128k$0.06290.56
Gemini 1.5 Pro (May)GoogleProprietary3466%37%24%2m$2.1968.10.37
Qwen TurboAlibabaProprietary3463%41%16%1m$0.09108.31.1
Llama 3.2 90B (Vision)MetaOpen3367%43%21%128k$0.8130.20.41
Qwen2 72BAlibabaOpen3362%37%16%131k$0.0031.11.32
Nova LiteAmazonProprietary3359%43%17%300k$0.10283.90.31
Gemini 1.5 Flash-8BGoogleProprietary3157%36%22%1m$0.07273.50.18
DeepHermes 3 - Mistral 24BNous ResearchOpen3058%38%20%32k$0.00
Jamba 1.5 LargeAI21 LabsOpen2957%43%14%256k$3.5067.20.53
Hermes 3 - Llama-3.1 70BNous ResearchOpen2957%40%19%128k$0.00
Jamba 1.6 LargeAI21 LabsOpen2956%39%17%256k$3.5060.80.53
Gemini 1.5 Flash (May)GoogleProprietary2857%32%20%1m$0.13311.90.27
Nova MicroAmazonProprietary2853%36%14%130k$0.06337.10.3
Yi-Large01.AIProprietary2859%36%11%32k$3.0068.30.47
Claude 3 SonnetAnthropicProprietary2858%40%18%200k$6.0061.30.61
Codestral (Jan '25)MistralProprietary2845%31%24%256k$0.45190.30.29
Llama 3 70BMetaOpen2757%38%20%8k$0.8846.30.46
Mistral Small (Sep '24)MistralOpen2753%38%14%33k$0.3065.30.33
Phi-4 MultimodalMicrosoft AzureOpen2749%32%13%128k$0.0021.90.34
Qwen2.5 Coder 7BAlibabaOpen2747%34%13%131k$0.032000.49
Mistral Large (Feb '24)MistralProprietary2652%35%18%33k$6.0030.30.48
Mixtral 8x22BMistralOpen2654%33%15%65k$3.0056.90.35
Phi-4 MiniMicrosoft AzureOpen2647%33%13%128k$0.0057.70.33
Phi-3 Medium 14BMicrosoft AzureOpen2554%33%15%128k$0.3052.20.41
Gemma 3 4BGoogleOpen2442%29%7%128k$0.031480.22
Claude 2.1AnthropicProprietary2450%32%20%200k$12.0013.90.88
Llama 3.1 8BMetaOpen2448%26%12%128k$0.10188.40.35
Pixtral 12BMistralOpen2347%34%12%128k$0.15102.20.3
Qwen3 0.6B (Reasoning)AlibabaOpen2335%24%12%32k$0.00
Mistral Small (Feb '24)MistralProprietary2342%30%11%33k$1.50153.60.27
Mistral MediumMistralProprietary2349%35%10%33k$4.0941.40.37
Ministral 8BMistralOpen2239%28%11%128k$0.10133.50.32
Gemma 2 9BGoogleOpen2250%31%13%8k$0.12
Phi-3 MiniMicrosoft AzureOpen2244%32%12%4k$0.00
LFM 40BLiquid AIProprietary2243%33%10%32k$0.15163.40.17
Command-R+CohereOpen2143%34%11%128k$4.3849.20.27
Llama 3 8BMetaOpen2141%30%10%8k$0.09103.30.34
Gemini 1.0 ProGoogleProprietary2143%28%12%33k$0.75
Codestral (May '24)MistralOpen2033%26%21%33k$0.30106.80.33
Aya Expanse 32BCohereOpen2038%23%14%128k$0.75121.50.16
Llama 2 Chat 13BMetaOpen2041%32%10%4k$0.00
Command-R+ (Apr '24)CohereOpen2043%32%12%128k$6.0066.90.24
DBRXDatabricksOpen2040%33%9%33k$1.13
Ministral 3BMistralProprietary2034%26%7%128k$0.04225.90.26
Mistral NeMoMistralOpen2040%31%6%128k$0.15143.20.29
Llama 3.2 3BMetaOpen2035%26%8%128k$0.05130.60.37
DeepSeek R1 Distill Qwen 1.5BDeepSeekOpen1927%10%7%128k$0.18379.95.51
Jamba 1.5 MiniAI21 LabsOpen1837%30%6%256k$0.251790.33
Jamba 1.6 MiniAI21 LabsOpen1837%30%7%256k$0.25195.40.33
Mixtral 8x7BMistralOpen1739%29%7%33k$0.7081.70.34
DeepHermes 3 - Llama-3.1 8BNous ResearchOpen1637%27%9%128k$0.00
Aya Expanse 8BCohereOpen1631%25%7%8k$0.75167.10.12
Command-RCohereOpen1534%29%4%128k$0.2673.80.2
Command-R (Mar '24)CohereOpen1534%28%5%128k$0.75166.20.14
Codestral-MambaMistralOpen1421%21%13%256k$0.2594.50.43
Mistral 7BMistralOpen1025%18%5%8k$0.25107.70.32
Llama 3.2 1BMetaOpen1020%20%2%128k$0.03155.50.44
Llama 2 Chat 7BMetaOpen816%23%0%4k$0.101330.39
GPT-4o mini Realtime (Dec '24)OpenAIProprietary128k$0.00
o1-proOpenAIProprietary200k$262.50
GPT-4o (ChatGPT)OpenAIProprietary77%51%128k$7.50
GPT-4o Realtime (Dec '24)OpenAIProprietary128k$0.00
Llama 3.2 11B (Vision)MetaOpen46%22%11%128k$0.1682.70.44
MiMo 7B RLXiaomiOpen33k$0.00
Gemma 3 1BGoogleOpen10%19%1%32k$0.00
Mistral SabaMistralProprietary61%42%32k$0.3091.30.33
Sonar Reasoning ProPerplexityProprietary127k$0.00
R1 1776PerplexityOpen128k$3.50
Sonar ReasoningPerplexityProprietary62%127k$2.0086.924.54
Grok 3 Reasoning BetaxAIProprietary1m$0.00
Grok 3 mini Reasoning (low)xAIProprietary1m$0.35110.618.33
Reka FlashReka AIProprietary128k$0.3533.40.95
Reka CoreReka AIProprietary128k$2.0027.70.96
Reka Flash (Feb '24)Reka AIProprietary128k$0.3545.90.93
Reka EdgeReka AIProprietary128k$0.1086.50.85
ArcticSnowflakeOpen4k$0.00
Qwen3 8B (Reasoning)AlibabaOpen74%59%26%128k$0.00
o1-previewOpenAIProprietary128k$26.25160.419.13
GPT-4o (Aug '24)OpenAIProprietary52%32%128k$4.3889.70.55
GPT-4 TurboOpenAIProprietary69%29%128k$15.0046.50.66
GPT-3.5 TurboOpenAIProprietary46%30%4k$0.75125.90.4
GPT-4.5 (Preview)OpenAIProprietary71%128k$93.75701.01
GPT-4OpenAIProprietary8k$37.5030.70.71
Llama 2 Chat 70BMetaOpen41%33%10%4k$0.00
Gemini 2.0 Flash-Lite (Preview)GoogleProprietary54%18%1m$0.13212.40.28
Gemma 2 27BGoogleOpen57%36%8k$0.26
Gemini 1.0 UltraGoogleProprietary33k$0.00
Gemini 2.0 Flash Thinking exp. (Dec '24)GoogleProprietary2m$0.00
Claude 3.5 Sonnet (June)AnthropicProprietary75%56%200k$6.0080.80.82
Claude 3 HaikuAnthropicProprietary15%200k$0.50137.90.5
Claude InstantAnthropicProprietary43%33%11%100k$1.2057.70.52
Claude 2.0AnthropicProprietary49%34%17%100k$12.0030.80.86
Llama 65BMetaOpen2k$0.00
DeepSeek-V2.5 (Dec '24)DeepSeekOpen128k$0.17
DeepSeek-Coder-V2DeepSeekOpen128k$0.17
DeepSeek LLM 67B (V1)DeepSeekOpen4k$0.00
DeepSeek-V2.5DeepSeekOpen128k$0.17
DeepSeek Coder V2 LiteDeepSeekOpen43%32%16%128k$0.09109.30.62
DeepSeek-V2DeepSeekOpen128k$0.17
OpenChat 3.5OpenChatOpen31%23%12%8k$0.0648.30.53
Solar MiniUpstageOpen4k$0.1513.61.76
Jamba InstructAI21 LabsProprietary34%27%5%256k$0.55175.60.33
Qwen1.5 Chat 110BAlibabaOpen29%32k$0.0023.71.55
Qwen Chat 72BAlibabaOpen34k$1.00