Model rankings
277 models · 2 boards · 2 sources
Smartest
Best open model
Best from China
Top three open models
DeepSeek R1DeepSeek · #3 overall60.2
DeepSeek R1 Distill Qwen 32BDeepSeek · #6 overall51.9
DeepSeek R1 Distill Qwen 14BDeepSeek · #8 overall49.0
This week's changes
SonarIntelligence · #15New
Sonar ProIntelligence · #16New
Nova LiteIntelligence · #42New
Jamba 1.6 LargeIntelligence · #45New
Phi-4 Multimodal InstructIntelligence · #62New
| Rank | Change | Model | Intelligence | Price $/M | Speed tok/s | First token |
|---|---|---|---|---|---|---|
| 1 | — | DeepSeek R1OpenDeepSeek | 60.2 | — | — | — |
| 2 | — | DeepSeek R1 Distill Qwen 32BOpenDeepSeek | 51.9 | — | — | — |
| 3 | — | DeepSeek R1 Distill Qwen 14BOpenDeepSeek | 49.0 | — | — | — |
| 4 | — | DeepSeek R1 Distill Llama 70BOpenDeepSeek | 48.2 | — | — | — |
| 5 | — | DeepSeek V3OpenDeepSeek | 45.7 | — | — | — |
| 6 | — | QwQ 32B-PreviewOpenAlibaba | 42.6 | — | — | — |
| 7 | — | Llama 3.3 Instruct 70BOpenMeta | 41.1 | — | — | — |
| 8 | — | Llama 3.1 Instruct 405BOpenMeta | 40.5 | — | — | — |
| 9 | — | Qwen2.5 Instruct 72BOpenAlibaba | 40.3 | — | — | — |
| 10 | — | Phi-4OpenMicrosoft Azure | 40.2 | — | — | — |
| 11 | — | Llama 3.1 Tulu3 405BOpenAllen Institute for AI | 39.9 | — | — | — |
| 12 | — | MiniMax-Text-01OpenMiniMax | 39.9 | — | — | — |
| 13 | — | Mistral Large 2 (Nov '24)OpenMistral · 1 more variant | 38.3 | — | — | — |
| 14 | — | Pixtral LargeOpenMistral | 37.4 | — | — | — |
| 15 | — | Qwen2.5 Instruct 32BOpenAlibaba | 37.4 | — | — | — |
| 16 | — | Llama 3.1 Nemotron Instruct 70BOpenNVIDIA | 37.3 | — | — | — |
| 17 | — | Qwen2.5 Coder Instruct 32BOpenAlibaba | 36.3 | — | — | — |
| 18 | — | Llama 3.1 Instruct 70BOpenMeta | 35.4 | — | — | — |
| 19 | — | Mistral Small 3OpenMistral | 35.3 | — | — | — |
| 20 | — | DeepSeek R1 Distill Llama 8BOpenDeepSeek | 34.0 | — | — | — |
| 21 | — | Llama 3.2 Instruct 90B (Vision)OpenMeta | 33.4 | — | — | — |
| 22 | — | Qwen2 Instruct 72BOpenAlibaba | 32.6 | — | — | — |
| 23 | — | Jamba 1.5 LargeOpenAI21 Labs | 29.3 | — | — | — |
| 24 | New | Jamba 1.6 LargeOpenAI21 Labs | 28.8 | — | — | — |
| 25 | ▼ 1 | Llama 3 Instruct 70BOpenMeta | 27.5 | — | — | — |
| 26 | ▼ 1 | Mistral Small (Sep '24)OpenMistral | 27.4 | — | — | — |
| 27 | ▼ 1 | Mixtral 8x22B InstructOpenMistral | 26.2 | — | — | — |
| 28 | ▼ 1 | Qwen2.5 Coder Instruct 7BOpenAlibaba | 26.1 | — | — | — |
| 29 | ▼ 1 | Phi-3 Medium Instruct 14BOpenMicrosoft Azure | 24.5 | — | — | — |
| 30 | ▼ 1 | DeepSeek Coder V2 Lite InstructOpenDeepSeek | 24.1 | — | — | — |
| 31 | ▼ 1 | Llama 3.1 Instruct 8BOpenMeta | 23.7 | — | — | — |
| 32 | ▼ 1 | Pixtral 12B (2409)OpenMistral | 23.4 | — | — | — |
| 33 | New | Phi-4 Multimodal InstructOpenMicrosoft Azure | 23.0 | — | — | — |
| 34 | New | Phi-4 Mini InstructOpenMicrosoft Azure | 22.8 | — | — | — |
| 35 | ▼ 3 | Ministral 8BOpenMistral | 22.3 | — | — | — |
| 36 | ▼ 3 | Llama 3.2 Instruct 11B (Vision)OpenMeta | 22.2 | — | — | — |
| 37 | ▼ 3 | Command-R+ (Aug '24)OpenCohere · 3 more variants | 21.5 | — | — | — |
| 38 | New | Llama 3 Instruct 8BOpenMeta | 21.5 | — | — | — |
| 39 | ▼ 4 | Codestral (May '24)OpenMistral | 20.1 | — | — | — |
| 40 | ▼ 4 | Aya Expanse 32BOpenCohere | 20.1 | — | — | — |
| 41 | ▼ 4 | DBRX InstructOpenDatabricks | 19.8 | — | — | — |
| 42 | ▼ 4 | Mistral NeMoOpenMistral | 19.6 | — | — | — |
| 43 | ▼ 4 | Llama 3.2 Instruct 3BOpenMeta | 19.5 | — | — | — |
| 44 | ▼ 4 | DeepSeek R1 Distill Qwen 1.5BOpenDeepSeek | 19.2 | — | — | — |
| 45 | New | Jamba 1.5 MiniOpenAI21 Labs | 18.4 | — | — | — |
| 46 | New | Jamba 1.6 MiniOpenAI21 Labs | 17.7 | — | — | — |
| 47 | ▼ 6 | Mixtral 8x7B InstructOpenMistral | 17.0 | — | — | — |
| 48 | New | Aya Expanse 8BOpenCohere | 16.5 | — | — | — |
| 49 | ▼ 7 | OpenChat 3.5 (1210)OpenOpenChat | 16.3 | — | — | — |
| 50 | ▼ 7 | Codestral-MambaOpenMistral | 13.9 | — | — | — |
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