模型排行
398 个模型 · 2 个榜单 · 2 个数据来源
智力王座
开源之光
性价比之王
国产之光
开源前三甲
DeepSeek R1 0528 (May '25)DeepSeek · 总第 5 名68.3
Qwen3 235B A22B (Reasoning)Alibaba · 总第 10 名62.3
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA · 总第 12 名60.8
这一期的变化
o3-pro智力榜 · 第 1新上榜
DeepSeek R1 0528 Qwen3 8B智力榜 · 第 27新上榜
Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)智力榜 · 第 43新上榜
| 排名 | 变化 | 模型 | 智力 | 价格 $/M | 速度 tok/s | 首字延迟 |
|---|---|---|---|---|---|---|
| 1 | 新 | o3-pro闭源OpenAI | 71.0 | $35 | — | — |
| 2 | ▲ 1 | Gemini 2.5 Pro Preview (Jun '25)闭源Google · 另有 2 个档位 | 70.5 | $3.4 | — | — |
| 3 | ▼ 2 | o3闭源OpenAI | 70.0 | $3.5 | — | — |
| 4 | ▼ 2 | o4-mini (high)闭源OpenAI | 69.8 | $1.9 | — | — |
| 5 | ▼ 1 | Grok 3 mini Reasoning (high)闭源xAI | 66.7 | $0.35 | — | — |
| 6 | ▼ 1 | o3-mini (high)闭源OpenAI · 另有 1 个档位 | 65.5 | $1.9 | — | — |
| 7 | ▼ 1 | Gemini 2.5 Flash Preview (May '25) (Reasoning)闭源Google · 另有 3 个档位 | 65.1 | $0.99 | — | — |
| 8 | ▼ 1 | Claude 4 Opus (Extended Thinking)闭源Anthropic · 另有 1 个档位 | 64.4 | $30 | — | — |
| 9 | ▼ 1 | o1闭源OpenAI · 另有 1 个档位 | 61.9 | $26 | — | — |
| 10 | ▼ 1 | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)开源NVIDIA | 60.8 | $0.90 | — | — |
| 11 | ▼ 1 | Claude 4 Sonnet (Extended Thinking)闭源Anthropic · 另有 1 个档位 | 60.8 | $6.0 | — | — |
| 12 | ▼ 1 | Claude 3.7 Sonnet (Extended Thinking)闭源Anthropic · 另有 1 个档位 | 57.4 | $6.0 | — | — |
| 13 | ▼ 1 | Grok 3 Reasoning Beta闭源xAI | 56.1 | $0.000 | — | — |
| 13 | ▼ 1 | o1-pro闭源OpenAI | 56.1 | $263 | — | — |
| 15 | ▼ 1 | o1-mini闭源OpenAI | 53.8 | $1.9 | — | — |
| 16 | ▼ 1 | GPT-4.5 (Preview)闭源OpenAI | 53.0 | $94 | — | — |
| 17 | ▼ 1 | GPT-4.1 mini闭源OpenAI | 52.9 | $0.70 | — | — |
| 18 | ▼ 1 | GPT-4.1闭源OpenAI | 52.6 | $3.5 | — | — |
| 19 | ▼ 1 | Gemini 2.0 Flash Thinking Experimental (Jan '25)闭源Google · 另有 1 个档位 | 52.3 | $0.000 | — | — |
| 20 | ▼ 1 | Llama 3.3 Nemotron Super 49B v1 (Reasoning)开源NVIDIA · 另有 1 个档位 | 51.2 | $0.000 | — | — |
| 21 | ▼ 1 | Grok 3闭源xAI | 50.6 | $6.0 | — | — |
| 22 | ▼ 1 | Llama 4 Maverick开源Meta | 50.5 | $0.38 | — | — |
| 23 | ▼ 1 | GPT-4o (March 2025, chatgpt-4o-latest)闭源OpenAI · 另有 4 个档位 | 50.3 | $7.5 | — | — |
| 24 | ▼ 1 | Gemini 2.0 Pro Experimental (Feb '25)闭源Google | 49.2 | $0.000 | — | — |
| 25 | ▼ 1 | Sonar Reasoning闭源Perplexity | 48.8 | $2.0 | — | — |
| 26 | ▼ 1 | Gemini 2.0 Flash (Feb '25)闭源Google · 另有 1 个档位 | 48.1 | $0.18 | — | — |
| 27 | 新 | Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)开源NVIDIA | 45.0 | $0.000 | — | — |
| 28 | ▼ 2 | Gemini 1.5 Pro (Sep '24)闭源Google · 另有 1 个档位 | 44.6 | $2.2 | — | — |
| 29 | ▼ 2 | Claude 3.5 Sonnet (Oct '24)闭源Anthropic · 另有 1 个档位 | 44.4 | $6.0 | — | — |
| 30 | ▼ 2 | Sonar闭源Perplexity | 43.4 | $1.0 | — | — |
| 31 | ▼ 2 | Llama 4 Scout开源Meta | 43.0 | $0.24 | — | — |
| 32 | ▼ 2 | Sonar Pro闭源Perplexity | 42.7 | $6.0 | — | — |
| 33 | ▼ 2 | Nova Premier闭源Amazon | 42.6 | $5.0 | — | — |
| 34 | ▼ 2 | Gemini 2.0 Flash-Lite (Feb '25)闭源Google · 另有 1 个档位 | 41.4 | $0.13 | — | — |
| 35 | ▼ 2 | Llama 3.3 Instruct 70B开源Meta | 41.1 | $0.62 | — | — |
| 36 | ▼ 2 | GPT-4.1 nano闭源OpenAI | 41.0 | $0.18 | — | — |
| 37 | ▼ 2 | Llama 3.1 Instruct 405B开源Meta | 40.5 | $3.5 | — | — |
| 38 | ▼ 2 | Phi-4开源Microsoft Azure | 40.2 | $0.22 | — | — |
| 39 | ▼ 2 | Grok 2 (Dec '24)闭源xAI | 39.2 | $0.000 | — | — |
| 40 | ▼ 2 | Gemini 1.5 Flash (Sep '24)闭源Google · 另有 1 个档位 | 39.0 | $0.13 | — | — |
| 41 | ▼ 2 | GPT-4 Turbo闭源OpenAI | 38.8 | $15 | — | — |
| 42 | ▼ 2 | Gemma 3 27B Instruct开源Google | 37.6 | $0.000 | — | — |
| 43 | ▼ 2 | Grok Beta闭源xAI | 37.5 | $7.5 | — | — |
| 44 | ▼ 2 | Llama 3.1 Nemotron Instruct 70B开源NVIDIA | 37.3 | $0.16 | — | — |
| 45 | ▼ 2 | Nova Pro闭源Amazon | 37.1 | $1.4 | — | — |
| 46 | ▼ 2 | GPT-4闭源OpenAI | 36.0 | $38 | — | — |
| 47 | ▼ 2 | GPT-4o mini闭源OpenAI | 35.7 | $0.26 | — | — |
| 48 | ▼ 2 | Llama 3.1 Instruct 70B开源Meta | 35.4 | $0.72 | — | — |
| 49 | ▼ 2 | Claude 3 Opus闭源Anthropic | 35.1 | $30 | — | — |
| 50 | ▼ 2 | Claude 3.5 Haiku闭源Anthropic | 34.7 | $1.6 | — | — |
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