模型排行
521 個模型 · 3 個榜單 · 2 個數據來源
智力王座
開源之光
價效比之王
國產之光
開源前三甲
Qwen3 235B A22B 2507 (Reasoning)Alibaba · 總第 8 名63.6
gpt-oss-120BOpenAI · 總第 9 名61.3
DeepSeek V3.1 (Reasoning)DeepSeek · 總第 11 名59.7
這一期的變化
Exaone 4.0 1.2B (Reasoning)智力榜 · 第 81新上榜
Qwen3 4B 2507 (Reasoning)智力榜 · 第 26上升 59 位
Qwen3 0.6B (Reasoning)智力榜 · 第 148上升 10 位
| 排名 | 變化 | 模型 | 智力 | 價格 $/M | 速度 tok/s | 首字延遲 |
|---|---|---|---|---|---|---|
| 1 | — | GPT-5 (high)閉源OpenAI · 另有 3 個檔位 | 69.0 | $3.4 | — | — |
| 2 | — | Grok 4閉源xAI | 67.5 | $6.0 | — | — |
| 3 | — | o3-pro閉源OpenAI | 67.5 | $35 | — | — |
| 4 | — | o3閉源OpenAI | 67.1 | $3.5 | — | — |
| 5 | — | GPT-5 mini (high)閉源OpenAI · 另有 2 個檔位 | 65.4 | $0.69 | — | — |
| 6 | — | o4-mini (high)閉源OpenAI | 65.1 | $1.9 | — | — |
| 7 | — | Gemini 2.5 Pro閉源Google | 64.6 | $3.4 | — | — |
| 8 | — | gpt-oss-120B開源OpenAI | 61.3 | $0.26 | — | — |
| 9 | — | Claude 4.1 Opus (Extended Thinking)閉源Anthropic · 另有 1 個檔位 | 61.0 | $30 | — | — |
| 10 | — | Gemini 2.5 Pro Preview (Mar' 25)閉源Google · 另有 1 個檔位 | 59.0 | $3.4 | — | — |
| 11 | — | Claude 4 Sonnet (Extended Thinking)閉源Anthropic · 另有 1 個檔位 | 58.8 | $6.0 | — | — |
| 12 | — | Gemini 2.5 Flash (Reasoning)閉源Google · 另有 1 個檔位 | 58.4 | $0.85 | — | — |
| 13 | — | Grok 3 mini Reasoning (high)閉源xAI | 58.0 | $0.35 | — | — |
| 14 | — | o3-mini (high)閉源OpenAI · 另有 1 個檔位 | 55.5 | $1.9 | — | — |
| 15 | — | Claude 4 Opus (Extended Thinking)閉源Anthropic · 另有 1 個檔位 | 55.3 | $30 | — | — |
| 16 | — | GPT-5 nano (high)閉源OpenAI · 另有 2 個檔位 | 54.7 | $0.14 | — | — |
| 17 | — | Llama Nemotron Super 49B v1.5 (Reasoning)開源NVIDIA · 另有 1 個檔位 | 52.4 | $0.000 | — | — |
| 18 | — | o1閉源OpenAI · 另有 1 個檔位 | 51.7 | $26 | — | — |
| 19 | — | Sonar Reasoning Pro閉源Perplexity | 50.7 | $0.000 | — | — |
| 20 | — | Gemini 2.5 Flash Preview (Reasoning)閉源Google · 另有 1 個檔位 | 50.2 | $0.99 | — | — |
| 21 | — | gpt-oss-20B開源OpenAI | 49.0 | $0.088 | — | — |
| 22 | — | Gemini 2.5 Flash-Lite (Reasoning)閉源Google · 另有 1 個檔位 | 47.3 | $0.18 | — | — |
| 23 | — | Claude 3.7 Sonnet (Extended Thinking)閉源Anthropic · 另有 1 個檔位 | 47.0 | $6.0 | — | — |
| 24 | — | GPT-4.1閉源OpenAI | 46.8 | $3.5 | — | — |
| 25 | — | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)開源NVIDIA | 46.4 | $0.90 | — | — |
| 26 | — | Grok 3 Reasoning Beta閉源xAI | 45.6 | $0.000 | — | — |
| 26 | — | o1-pro閉源OpenAI | 45.6 | $263 | — | — |
| 28 | — | GPT-4.1 mini閉源OpenAI | 45.6 | $0.70 | — | — |
| 29 | — | o1-mini閉源OpenAI | 43.3 | $1.9 | — | — |
| 30 | — | NVIDIA Nemotron Nano 9B V2 (Reasoning)開源NVIDIA · 另有 1 個檔位 | 43.2 | $0.000 | — | — |
| 31 | — | GPT-4.5 (Preview)閉源OpenAI | 42.4 | $0.000 | — | — |
| 32 | — | Llama 4 Maverick開源Meta | 41.7 | $0.39 | — | — |
| 33 | — | Gemini 2.0 Flash Thinking Experimental (Jan '25)閉源Google · 另有 1 個檔位 | 41.7 | $0.000 | — | — |
| 34 | — | Grok 3閉源xAI | 39.9 | $6.0 | — | — |
| 35 | — | GPT-4o (March 2025, chatgpt-4o-latest)閉源OpenAI · 另有 4 個檔位 | 39.5 | $7.5 | — | — |
| 36 | — | Llama 3.3 Nemotron Super 49B v1 (Reasoning)開源NVIDIA · 另有 1 個檔位 | 39.4 | $0.000 | — | — |
| 37 | — | Gemini 2.0 Pro Experimental (Feb '25)閉源Google | 38.4 | $0.000 | — | — |
| 38 | — | Sonar Reasoning閉源Perplexity | 38.0 | $2.0 | — | — |
| 39 | — | Gemini 2.0 Flash (Feb '25)閉源Google · 另有 1 個檔位 | 37.8 | $0.18 | — | — |
| 40 | — | Nova Premier閉源Amazon | 34.8 | $5.0 | — | — |
| 41 | ▲ 1 | Gemini 1.5 Pro (Sep '24)閉源Google · 另有 1 個檔位 | 33.6 | $0.000 | — | — |
| 42 | ▲ 1 | Claude 3.5 Sonnet (Oct '24)閉源Anthropic · 另有 1 個檔位 | 33.5 | $6.0 | — | — |
| 43 | ▲ 1 | Llama 4 Scout開源Meta | 33.1 | $0.23 | — | — |
| 44 | ▲ 1 | Sonar閉源Perplexity | 32.4 | $1.0 | — | — |
| 45 | ▲ 1 | Sonar Pro閉源Perplexity | 31.7 | $6.0 | — | — |
| 46 | ▲ 1 | GPT-4.1 nano閉源OpenAI | 31.6 | $0.18 | — | — |
| 47 | ▲ 1 | Llama 3.3 Instruct 70B開源Meta | 31.2 | $0.57 | — | — |
| 48 | ▲ 1 | Gemini 2.0 Flash-Lite (Feb '25)閉源Google · 另有 1 個檔位 | 30.3 | $0.13 | — | — |
| 49 | ▼ 8 | Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)開源NVIDIA | 29.6 | $0.000 | — | — |
| 50 | — | Llama 3.1 Instruct 405B開源Meta | 28.9 | $3.5 | — | — |
智力進步
漲得最多 · Anthropic +57.2目前最高 · Anthropic 64.6最高與最低相差 18.1 分時間範圍 2025年11月 – 2026年11月
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi