LLM 榜单
更新于: 2026-10-20| # | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAI | 96% | 54.7% | 89.9% | — | — | — | 80.2 | $30.00 |
| 2 | MiniMax M3.1 FlashMiniMax | — | — | — | 73.7% | — | — | 73.7 | $0.375 |
| 3 | Claude Opus 5Anthropic | 93.4% | 56.3% | 89.1% | 74% | 50.3% | 75.3% | 73.1 | $15.00 |
| 4 | Grok 4.7xAI | — | — | — | 71% | — | — | 71.0 | $4.00 |
| 5 | DeepSeek V4.1-Flash 0910DeepSeek | 90.9% | 36.8% | 90.6% | 74.2% | 54.8% | 65.4% | 68.8 | $0.750 |
| 6 | GPT 5.6-SolOpenAI | 94.1% | 44.5% | 88.8% | 73% | 45.8% | 56.8% | 67.2 | $12.00 |
| 7 | Kimi K3月之暗面 | 92.9% | 43.5% | 88.3% | 67.5% | 46.7% | 58% | 66.1 | — |
| 8 | GLM 5.3智谱 | 88.1% | 42% | 88.2% | 66.9% | 48.8% | 58% | 65.3 | — |
| 9 | DeepSeek V4-Pro 0813DeepSeek | 92.4% | 42.7% | 87.9% | 62.7% | 43.2% | 61.5% | 65.1 | $2.64 |
| 10 | Gemini 4 ArgonGoogle | — | — | — | 77.9% | 51.3% | — | 64.6 | $6.00 |
| 11 | Claude Opus 5.5Anthropic | — | 64.4% | — | — | — | — | 64.4 | $12.00 |
| 12 | DeepSeek V4.1-Flash 1017DeepSeek | — | 42.6% | — | 77% | 60.5% | 68.2% | 62.1 | $0.750 |
| 13 | DeepSeek V4-Flash 0731DeepSeek | 89.9% | 37.8% | 82.7% | 54.4% | 37.7% | 54.2% | 59.4 | — |
| 14 | Gemini 3.1 ProGoogle | 94.3% | 44.4% | — | — | 35.4% | — | 58.0 | $7.00 |
| 15 | MiniMax M3MiniMax | — | — | 66% | — | — | 42.1% | 54.0 | $0.750 |
这里的每个分数都能追溯到具名来源，每个价格都能追溯到厂商页面。不插值、不画营销图——只给你站得住脚的数字。
热门对比
Claude Opus 5 vs GPT 5.6-SolClaude Opus 5 vs Gemini 3.1 ProClaude Opus 5 vs DeepSeek V4.1-Flash 0910GPT 5.6-Sol vs Gemini 3.1 ProGPT 5.6-Sol vs DeepSeek V4.1-Flash 0910Gemini 3.1 Pro vs Grok 4.7Gemini 3.1 Pro vs DeepSeek V4.1-Flash 0910Grok 4.7 vs Claude Opus 5Grok 4.7 vs DeepSeek V4.1-Flash 0910Claude Opus 5 vs GLM 5.3GPT 5.6-Sol vs Kimi K3Gemini 3.1 Pro vs GLM 5.3
新模型发布
每一次大模型发布，都重写成一篇成本与换型决策分析。页面上的价格和跑分都是从厂商官方页面实时读取的，不是抄进文章里的。