模型排行
347 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
Qwen3 235B A22B (Reasoning)Alibaba · 总第 6 名62.3
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA · 总第 8 名60.8
DeepSeek R1DeepSeek · 总第 10 名60.2
这一期的变化
Qwen2.5 Turbo智力榜 · 第 70新上榜
| 排名 | 变化 | 模型 | 智力 | 价格 $/M | 速度 tok/s | 首字延迟 |
|---|---|---|---|---|---|---|
| 1 | — | o4-mini (high)闭源OpenAI | 69.8 | — | — | — |
| 2 | — | Gemini 2.5 Pro Preview (Mar' 25)闭源Google | 68.9 | — | — | — |
| 3 | — | o3闭源OpenAI | 66.8 | — | — | — |
| 4 | — | Grok 3 mini Reasoning (high)闭源xAI | 66.7 | — | — | — |
| 5 | — | o3-mini (high)闭源OpenAI · 另有 1 个档位 | 65.5 | — | — | — |
| 6 | — | Qwen3 235B A22B (Reasoning)开源Alibaba · 另有 1 个档位 | 62.3 | — | — | — |
| 7 | — | o1闭源OpenAI | 61.9 | — | — | — |
| 8 | — | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)开源NVIDIA | 60.8 | — | — | — |
| 9 | — | Gemini 2.5 Flash Preview (Reasoning)闭源Google · 另有 1 个档位 | 60.5 | — | — | — |
| 10 | — | DeepSeek R1开源DeepSeek | 60.2 | — | — | — |
| 11 | — | Qwen3 32B (Reasoning)开源Alibaba · 另有 1 个档位 | 59.2 | — | — | — |
| 12 | — | QwQ 32B开源Alibaba | 58.1 | — | — | — |
| 13 | — | Claude 3.7 Sonnet (Extended Thinking)闭源Anthropic · 另有 1 个档位 | 57.4 | — | — | — |
| 14 | — | Qwen3 14B (Reasoning)开源Alibaba · 另有 1 个档位 | 55.7 | — | — | — |
| 15 | — | Qwen3 30B A3B (Reasoning)开源Alibaba · 另有 1 个档位 | 55.6 | — | — | — |
| 16 | — | o1-mini闭源OpenAI | 53.8 | — | — | — |
| 17 | — | DeepSeek V3 0324 (Mar' 25)开源DeepSeek · 另有 1 个档位 | 53.2 | — | — | — |
| 18 | — | GPT-4.1 mini闭源OpenAI | 52.9 | — | — | — |
| 19 | — | GPT-4.1闭源OpenAI | 52.6 | — | — | — |
| 20 | — | Gemini 2.0 Flash Thinking Experimental (Jan '25)闭源Google | 52.3 | — | — | — |
| 21 | — | DeepSeek R1 Distill Qwen 32B开源DeepSeek | 51.9 | — | — | — |
| 22 | — | Qwen3 8B (Reasoning)开源Alibaba · 另有 1 个档位 | 51.4 | — | — | — |
| 23 | — | Llama 3.3 Nemotron Super 49B v1 (Reasoning)开源NVIDIA · 另有 1 个档位 | 51.2 | — | — | — |
| 24 | — | Grok 3闭源xAI | 50.6 | — | — | — |
| 25 | — | Llama 4 Maverick开源Meta | 50.5 | — | — | — |
| 26 | — | GPT-4o (March 2025, chatgpt-4o-latest)闭源OpenAI · 另有 2 个档位 | 50.3 | — | — | — |
| 27 | — | Gemini 2.0 Pro Experimental (Feb '25)闭源Google | 49.2 | — | — | — |
| 28 | — | DeepSeek R1 Distill Qwen 14B开源DeepSeek | 49.0 | — | — | — |
| 29 | — | Mistral Medium 3闭源Mistral | 49.0 | — | — | — |
| 30 | — | DeepSeek R1 Distill Llama 70B开源DeepSeek | 48.2 | — | — | — |
| 31 | — | Gemini 2.0 Flash (Feb '25)闭源Google · 另有 1 个档位 | 48.1 | — | — | — |
| 32 | — | Qwen3 4B (Reasoning)开源Alibaba · 另有 1 个档位 | 47.3 | — | — | — |
| 33 | — | Reka Flash 3开源Reka AI | 47.1 | — | — | — |
| 34 | — | Qwen2.5 Max闭源Alibaba | 45.3 | — | — | — |
| 35 | — | Gemini 1.5 Pro (Sep '24)闭源Google · 另有 1 个档位 | 44.6 | — | — | — |
| 36 | — | Claude 3.5 Sonnet (Oct '24)闭源Anthropic | 44.4 | — | — | — |
| 37 | — | Sonar闭源Perplexity | 43.4 | — | — | — |
| 38 | — | Llama 4 Scout开源Meta | 43.0 | — | — | — |
| 39 | — | Sonar Pro闭源Perplexity | 42.7 | — | — | — |
| 40 | — | QwQ 32B-Preview开源Alibaba | 42.6 | — | — | — |
| 41 | — | Nova Premier闭源Amazon | 42.6 | — | — | — |
| 42 | — | Gemini 2.0 Flash-Lite (Feb '25)闭源Google | 41.4 | — | — | — |
| 43 | — | Llama 3.3 Instruct 70B开源Meta | 41.1 | — | — | — |
| 44 | — | GPT-4.1 nano闭源OpenAI | 41.0 | — | — | — |
| 45 | — | Llama 3.1 Instruct 405B开源Meta | 40.5 | — | — | — |
| 46 | — | Qwen2.5 Instruct 72B开源Alibaba | 40.4 | — | — | — |
| 47 | — | MiniMax-Text-01开源MiniMax | 40.2 | — | — | — |
| 48 | — | Phi-4开源Microsoft Azure | 40.2 | — | — | — |
| 49 | — | Command A开源Cohere | 40.0 | — | — | — |
| 50 | — | Llama 3.1 Tulu3 405B开源Allen Institute for AI | 39.9 | — | — | — |
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