模型排行
315 个模型 · 2 个榜单 · 2 个数据来源
智力王座
开源之光
国产之光
开源前三甲
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA · 总第 7 名60.8
DeepSeek R1DeepSeek · 总第 9 名60.2
QwQ 32BAlibaba · 总第 10 名58.1
这一期的变化
o4-mini (high)智力榜 · 第 1新上榜
o3智力榜 · 第 3新上榜
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)智力榜 · 第 7新上榜
Gemini 2.5 Flash Preview (Reasoning)智力榜 · 第 8新上榜
| 排名 | 变化 | 模型 | 智力 | 价格 $/M | 速度 tok/s | 首字延迟 |
|---|---|---|---|---|---|---|
| 1 | 新 | o4-mini (high)闭源OpenAI | 69.8 | — | — | — |
| 2 | ▼ 1 | Gemini 2.5 Pro Preview (Mar' 25)闭源Google | 67.8 | — | — | — |
| 3 | 新 | o3闭源OpenAI | 66.8 | — | — | — |
| 4 | ▼ 2 | Grok 3 mini Reasoning (high)闭源xAI | 66.7 | — | — | — |
| 5 | ▼ 2 | o3-mini (high)闭源OpenAI | 65.5 | — | — | — |
| 6 | ▼ 2 | o3-mini闭源OpenAI | 62.9 | — | — | — |
| 7 | ▼ 2 | o1闭源OpenAI | 61.9 | — | — | — |
| 8 | 新 | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)开源NVIDIA | 60.8 | — | — | — |
| 9 | 新 | Gemini 2.5 Flash Preview (Reasoning)闭源Google | 60.5 | — | — | — |
| 10 | ▼ 4 | DeepSeek R1开源DeepSeek | 60.2 | — | — | — |
| 11 | ▼ 4 | QwQ 32B开源Alibaba | 58.1 | — | — | — |
| 12 | ▼ 4 | Claude 3.7 Sonnet (Extended Thinking)闭源Anthropic | 57.4 | — | — | — |
| 13 | ▼ 4 | o1-mini闭源OpenAI | 53.8 | — | — | — |
| 14 | ▼ 4 | DeepSeek V3 0324 (Mar' 25)开源DeepSeek | 53.2 | — | — | — |
| 15 | ▼ 4 | GPT-4.1 mini闭源OpenAI | 52.9 | — | — | — |
| 16 | ▼ 4 | GPT-4.1闭源OpenAI | 52.6 | — | — | — |
| 17 | ▼ 4 | Gemini 2.0 Flash Thinking Experimental (Jan '25)闭源Google | 52.3 | — | — | — |
| 18 | ▼ 4 | DeepSeek R1 Distill Qwen 32B开源DeepSeek | 51.9 | — | — | — |
| 19 | ▼ 4 | Grok 3闭源xAI | 50.6 | — | — | — |
| 20 | ▼ 4 | Llama 4 Maverick开源Meta | 50.5 | — | — | — |
| 21 | ▼ 4 | GPT-4o (March 2025, chatgpt-4o-latest)闭源OpenAI | 50.3 | — | — | — |
| 22 | ▼ 4 | Gemini 2.0 Pro Experimental (Feb '25)闭源Google | 49.2 | — | — | — |
| 23 | ▼ 4 | DeepSeek R1 Distill Qwen 14B开源DeepSeek | 49.0 | — | — | — |
| 24 | ▼ 4 | DeepSeek R1 Distill Llama 70B开源DeepSeek | 48.2 | — | — | — |
| 25 | ▼ 4 | Claude 3.7 Sonnet (Standard)闭源Anthropic | 48.2 | — | — | — |
| 26 | ▼ 4 | Gemini 2.0 Flash (Feb '25)闭源Google | 48.1 | — | — | — |
| 27 | ▼ 4 | Reka Flash 3开源Reka AI | 47.1 | — | — | — |
| 28 | ▼ 4 | Gemini 2.0 Flash (experimental)闭源Google | 46.4 | — | — | — |
| 29 | ▼ 4 | DeepSeek V3 (Dec '24)开源DeepSeek | 45.6 | — | — | — |
| 30 | ▼ 4 | Qwen2.5 Max闭源Alibaba | 45.3 | — | — | — |
| 31 | ▼ 4 | Gemini 1.5 Pro (Sep '24)闭源Google | 44.6 | — | — | — |
| 32 | ▼ 4 | Claude 3.5 Sonnet (Oct '24)闭源Anthropic | 44.4 | — | — | — |
| 33 | ▼ 4 | Sonar闭源Perplexity | 43.4 | — | — | — |
| 34 | ▼ 4 | Llama 4 Scout开源Meta | 43.0 | — | — | — |
| 35 | ▼ 4 | Sonar Pro闭源Perplexity | 42.7 | — | — | — |
| 36 | ▼ 4 | QwQ 32B-Preview开源Alibaba | 42.6 | — | — | — |
| 37 | ▼ 4 | GPT-4o (Nov '24)闭源OpenAI | 41.5 | — | — | — |
| 38 | ▼ 4 | Gemini 2.0 Flash-Lite (Feb '25)闭源Google | 41.4 | — | — | — |
| 39 | ▼ 4 | Llama 3.3 Instruct 70B开源Meta | 41.1 | — | — | — |
| 40 | ▼ 4 | GPT-4.1 nano闭源OpenAI | 41.0 | — | — | — |
| 41 | ▼ 4 | GPT-4o (May '24)闭源OpenAI | 40.9 | — | — | — |
| 42 | ▼ 4 | Llama 3.1 Instruct 405B开源Meta | 40.5 | — | — | — |
| 43 | ▼ 4 | Qwen2.5 Instruct 72B开源Alibaba | 40.4 | — | — | — |
| 44 | ▼ 4 | MiniMax-Text-01开源MiniMax | 40.2 | — | — | — |
| 45 | ▼ 4 | Phi-4开源Microsoft Azure | 40.2 | — | — | — |
| 46 | ▼ 4 | Command A开源Cohere | 40.0 | — | — | — |
| 47 | ▼ 4 | Llama 3.1 Tulu3 405B开源Allen Institute for AI | 39.9 | — | — | — |
| 48 | ▼ 4 | Llama 3.3 Nemotron Super 49B v1开源NVIDIA | 39.3 | — | — | — |
| 49 | ▼ 4 | Grok 2 (Dec '24)闭源xAI | 39.2 | — | — | — |
| 50 | ▼ 4 | Gemini 1.5 Flash (Sep '24)闭源Google | 39.0 | — | — | — |
智力进步
涨得最多 · Anthropic +57.2目前最高 · Anthropic 64.6最高与最低相差 18.1 分时间范围 2025年11月 – 2026年11月
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi