Model rankings
315 models · 2 boards · 2 sources
Smartest
Best open model
Best from China
Top three by Intelligence
o4-mini (high)OpenAI69.8
Gemini 2.5 Pro Preview (Mar' 25)Google67.8
o3OpenAI66.8
Top three open models
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA · #7 overall60.8
DeepSeek R1DeepSeek · #9 overall60.2
QwQ 32BAlibaba · #10 overall58.1
| Rank | Change | Model | Intelligence | Price $/M | Speed tok/s | First token |
|---|---|---|---|---|---|---|
| 1 | — | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA | 60.8 | — | — | — |
| 2 | — | DeepSeek R1OpenDeepSeek | 60.2 | — | — | — |
| 3 | — | QwQ 32BOpenAlibaba | 58.1 | — | — | — |
| 4 | — | DeepSeek V3 0324 (Mar' 25)OpenDeepSeek · 1 more variant | 53.2 | — | — | — |
| 5 | — | DeepSeek R1 Distill Qwen 32BOpenDeepSeek | 51.9 | — | — | — |
| 6 | — | Llama 4 MaverickOpenMeta | 50.5 | — | — | — |
| 7 | — | DeepSeek R1 Distill Qwen 14BOpenDeepSeek | 49.0 | — | — | — |
| 8 | — | DeepSeek R1 Distill Llama 70BOpenDeepSeek | 48.2 | — | — | — |
| 9 | — | Reka Flash 3OpenReka AI | 47.1 | — | — | — |
| 10 | — | Llama 4 ScoutOpenMeta | 43.0 | — | — | — |
| 11 | — | QwQ 32B-PreviewOpenAlibaba | 42.6 | — | — | — |
| 12 | — | Llama 3.3 Instruct 70BOpenMeta | 41.1 | — | — | — |
| 13 | — | Llama 3.1 Instruct 405BOpenMeta | 40.5 | — | — | — |
| 14 | — | Qwen2.5 Instruct 72BOpenAlibaba | 40.4 | — | — | — |
| 15 | — | MiniMax-Text-01OpenMiniMax | 40.2 | — | — | — |
| 16 | — | Phi-4OpenMicrosoft Azure | 40.2 | — | — | — |
| 17 | — | Command AOpenCohere | 40.0 | — | — | — |
| 18 | — | Llama 3.1 Tulu3 405BOpenAllen Institute for AI | 39.9 | — | — | — |
| 19 | — | Llama 3.3 Nemotron Super 49B v1OpenNVIDIA | 39.3 | — | — | — |
| 20 | — | Mistral Large 2 (Nov '24)OpenMistral · 1 more variant | 38.3 | — | — | — |
| 21 | — | Gemma 3 27B InstructOpenGoogle | 37.6 | — | — | — |
| 22 | — | Pixtral LargeOpenMistral | 37.4 | — | — | — |
| 23 | — | Qwen2.5 Instruct 32BOpenAlibaba | 37.4 | — | — | — |
| 24 | — | Llama 3.1 Nemotron Instruct 70BOpenNVIDIA | 37.3 | — | — | — |
| 25 | — | Qwen2.5 Coder Instruct 32BOpenAlibaba | 36.3 | — | — | — |
| 26 | — | Llama 3.1 Instruct 70BOpenMeta | 35.4 | — | — | — |
| 27 | — | Mistral Small 3.1OpenMistral | 35.3 | — | — | — |
| 28 | — | Mistral Small 3OpenMistral | 35.3 | — | — | — |
| 29 | — | DeepSeek R1 Distill Llama 8BOpenDeepSeek | 34.0 | — | — | — |
| 30 | — | Gemma 3 12B InstructOpenGoogle | 33.8 | — | — | — |
| 31 | — | Llama 3.2 Instruct 90B (Vision)OpenMeta | 33.4 | — | — | — |
| 32 | — | Qwen2 Instruct 72BOpenAlibaba | 32.6 | — | — | — |
| 33 | — | DeepHermes 3 - Mistral 24B PreviewOpenNous Research | 30.0 | — | — | — |
| 34 | — | Jamba 1.5 LargeOpenAI21 Labs | 29.3 | — | — | — |
| 35 | — | Hermes 3 - Llama-3.1 70BOpenNous Research | 29.1 | — | — | — |
| 36 | — | Jamba 1.6 LargeOpenAI21 Labs | 28.8 | — | — | — |
| 37 | — | Llama 3 Instruct 70BOpenMeta | 27.5 | — | — | — |
| 38 | — | Mistral Small (Sep '24)OpenMistral | 27.4 | — | — | — |
| 39 | — | Phi-4 Multimodal InstructOpenMicrosoft Azure | 26.9 | — | — | — |
| 40 | — | Qwen2.5 Coder Instruct 7BOpenAlibaba | 26.7 | — | — | — |
| 41 | — | Mixtral 8x22B InstructOpenMistral | 26.2 | — | — | — |
| 42 | — | Phi-4 Mini InstructOpenMicrosoft Azure | 26.0 | — | — | — |
| 43 | — | Phi-3 Medium Instruct 14BOpenMicrosoft Azure | 24.5 | — | — | — |
| 44 | — | Gemma 3 4B InstructOpenGoogle | 24.5 | — | — | — |
| 45 | — | Llama 3.1 Instruct 8BOpenMeta | 23.7 | — | — | — |
| 46 | — | Pixtral 12B (2409)OpenMistral | 23.4 | — | — | — |
| 47 | — | Ministral 8BOpenMistral | 22.3 | — | — | — |
| 48 | — | Gemma 2 9BOpenGoogle | 22.2 | — | — | — |
| 49 | — | Phi-3 Mini Instruct 3.8BOpenMicrosoft Azure | 22.1 | — | — | — |
| 50 | — | Command-R+ (Aug '24)OpenCohere · 3 more variants | 21.5 | — | — | — |
Intelligence over time
Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi