모델 순위
모델 307개 · 순위표 2개 · 데이터 출처 2곳
가장 똑똑한 모델
오픈 모델 1위
중국 모델 1위
지능 상위 3
Gemini 2.5 Pro Preview (Mar' 25)Google67.8
Grok 3 mini Reasoning (high)xAI66.7
o3-mini (high)OpenAI65.5
오픈 모델 상위 3
DeepSeek R1DeepSeek · 전체 5위60.2
QwQ 32BAlibaba · 전체 6위58.1
DeepSeek V3 0324 (Mar' 25)DeepSeek · 전체 9위53.2
이번 주 변동
Gemini 2.5 Pro Preview (Mar' 25)지능 · 1위신규
Grok 3 mini Reasoning (high)지능 · 2위신규
GPT-4.1 mini지능 · 10위신규
GPT-4.1지능 · 11위신규
Grok 3지능 · 14위신규
| 순위 | 변동 | 모델 | 지능 | 가격 $/M | 속도 tok/s | 첫 토큰 |
|---|---|---|---|---|---|---|
| 1 | — | DeepSeek R1오픈DeepSeek | 60.2 | — | — | — |
| 2 | — | QwQ 32B오픈Alibaba | 58.1 | — | — | — |
| 3 | — | DeepSeek V3 0324 (Mar' 25)오픈DeepSeek | 53.2 | — | — | — |
| 4 | — | DeepSeek R1 Distill Qwen 32B오픈DeepSeek | 51.9 | — | — | — |
| 5 | — | Llama 4 Maverick오픈Meta | 50.5 | — | — | — |
| 6 | — | DeepSeek R1 Distill Qwen 14B오픈DeepSeek | 49.0 | — | — | — |
| 7 | — | DeepSeek R1 Distill Llama 70B오픈DeepSeek | 48.2 | — | — | — |
| 8 | — | Reka Flash 3오픈Reka AI | 47.1 | — | — | — |
| 9 | — | DeepSeek V3 (Dec '24)오픈DeepSeek | 45.6 | — | — | — |
| 10 | — | Llama 4 Scout오픈Meta | 43.0 | — | — | — |
| 11 | — | QwQ 32B-Preview오픈Alibaba | 42.6 | — | — | — |
| 12 | — | Llama 3.3 Instruct 70B오픈Meta | 41.1 | — | — | — |
| 13 | — | Llama 3.1 Instruct 405B오픈Meta | 40.5 | — | — | — |
| 14 | — | Qwen2.5 Instruct 72B오픈Alibaba | 40.4 | — | — | — |
| 15 | — | MiniMax-Text-01오픈MiniMax | 40.2 | — | — | — |
| 16 | — | Phi-4오픈Microsoft Azure | 40.2 | — | — | — |
| 17 | — | Command A오픈Cohere | 40.0 | — | — | — |
| 18 | — | Llama 3.1 Tulu3 405B오픈Allen Institute for AI | 39.9 | — | — | — |
| 19 | — | Llama 3.3 Nemotron Super 49B v1오픈NVIDIA | 39.3 | — | — | — |
| 20 | — | Mistral Large 2 (Nov '24)오픈Mistral | 38.3 | — | — | — |
| 21 | — | Gemma 3 27B Instruct오픈Google | 37.6 | — | — | — |
| 22 | — | Pixtral Large오픈Mistral | 37.4 | — | — | — |
| 23 | — | Qwen2.5 Instruct 32B오픈Alibaba | 37.4 | — | — | — |
| 24 | — | Llama 3.1 Nemotron Instruct 70B오픈NVIDIA | 37.3 | — | — | — |
| 25 | — | Mistral Large 2 (Jul '24)오픈Mistral | 36.6 | — | — | — |
| 26 | — | Qwen2.5 Coder Instruct 32B오픈Alibaba | 36.3 | — | — | — |
| 27 | — | Llama 3.1 Instruct 70B오픈Meta | 35.4 | — | — | — |
| 28 | — | Mistral Small 3.1오픈Mistral | 35.3 | — | — | — |
| 29 | — | Mistral Small 3오픈Mistral | 35.3 | — | — | — |
| 30 | — | DeepSeek R1 Distill Llama 8B오픈DeepSeek | 34.0 | — | — | — |
| 31 | — | Gemma 3 12B Instruct오픈Google | 33.8 | — | — | — |
| 32 | — | Llama 3.2 Instruct 90B (Vision)오픈Meta | 33.4 | — | — | — |
| 33 | — | Qwen2 Instruct 72B오픈Alibaba | 32.6 | — | — | — |
| 34 | 신규 | DeepHermes 3 - Mistral 24B Preview오픈Nous Research | 30.0 | — | — | — |
| 35 | ▼ 1 | Jamba 1.5 Large오픈AI21 Labs | 29.3 | — | — | — |
| 36 | 신규 | Hermes 3 - Llama-3.1 70B오픈Nous Research | 29.1 | — | — | — |
| 37 | ▼ 2 | Jamba 1.6 Large오픈AI21 Labs | 28.8 | — | — | — |
| 38 | ▼ 2 | Llama 3 Instruct 70B오픈Meta | 27.5 | — | — | — |
| 39 | ▼ 2 | Mistral Small (Sep '24)오픈Mistral | 27.4 | — | — | — |
| 40 | ▼ 2 | Phi-4 Multimodal Instruct오픈Microsoft Azure | 26.9 | — | — | — |
| 41 | ▼ 2 | Qwen2.5 Coder Instruct 7B오픈Alibaba | 26.7 | — | — | — |
| 42 | ▼ 2 | Mixtral 8x22B Instruct오픈Mistral | 26.2 | — | — | — |
| 43 | ▼ 2 | Phi-4 Mini Instruct오픈Microsoft Azure | 26.0 | — | — | — |
| 44 | ▼ 2 | Phi-3 Medium Instruct 14B오픈Microsoft Azure | 24.5 | — | — | — |
| 45 | ▼ 2 | Gemma 3 4B Instruct오픈Google | 24.5 | — | — | — |
| 46 | ▼ 2 | Llama 3.1 Instruct 8B오픈Meta | 23.7 | — | — | — |
| 47 | ▼ 2 | Pixtral 12B (2409)오픈Mistral | 23.4 | — | — | — |
| 48 | ▼ 2 | Ministral 8B오픈Mistral | 22.3 | — | — | — |
| 49 | ▼ 2 | Gemma 2 9B오픈Google | 22.2 | — | — | — |
| 50 | ▼ 2 | Phi-3 Mini Instruct 3.8B오픈Microsoft Azure | 22.1 | — | — | — |
지능의 발전
가장 많이 오름 · Anthropic +57.2현재 최고 · Anthropic 64.6최고와 최저의 차이 18.1점기간 2025년 11월 – 2026년 11월
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi