模型排行
1055 个模型 · 5 个榜单 · 2 个数据来源
网页王座
开源之光
国产之光
最多人测
得分前三甲
claude-opus-4-6-thinkingAnthropic1,552
claude-sonnet-4-6Anthropic1,524
claude-opus-4-5-20251101-thinking-32kAnthropic1,493
这一期的变化
GPT-5.4 (xhigh)智力榜 · 第 2新上榜
gpt-5.4-high (codex-harness)编程榜 · 第 5新上榜
gpt-5.4-medium (codex-harness)编程榜 · 第 12新上榜
Step 3.5 Flash智力榜 · 第 35新上榜
Gemini 3.1 Flash-Lite Preview智力榜 · 第 46新上榜
| 排名 | 变化 | 模型 | 得分 | 置信区间 | 投票 | 价格 $/M |
|---|---|---|---|---|---|---|
| 1 | — | claude-opus-4-6-thinking闭源Anthropic · 另有 1 个档位 | 1,552 | ±13 | 2,891 | — |
| 2 | — | claude-sonnet-4-6闭源Anthropic | 1,524 | ±11 | 4,322 | — |
| 3 | — | claude-opus-4-5-20251101-thinking-32k闭源Anthropic | 1,493 | ±7 | 12,499 | — |
| 4 | — | claude-opus-4-5-20251101闭源Anthropic | 1,472 | ±7 | 12,651 | — |
| 5 | 新 | gpt-5.4-high (codex-harness)闭源OpenAI | 1,460 | ±17 | 1,410 | — |
| 6 | ▼ 1 | gemini-3.1-pro-preview闭源Google | 1,457 | ±11 | 3,510 | — |
| 7 | ▼ 1 | glm-5开源Z.ai | 1,447 | ±11 | 3,608 | — |
| 8 | ▲ 1 | glm-4.7开源Z.ai | 1,442 | ±10 | 5,136 | — |
| 9 | ▼ 1 | gemini-3-flash闭源Google · 另有 1 个档位 | 1,441 | ±7 | 13,639 | — |
| 10 | ▼ 3 | gemini-3-pro闭源Google | 1,440 | ±7 | 17,867 | — |
| 11 | ▼ 1 | kimi-k2.5-thinking开源Moonshot AI | 1,437 | ±9 | 5,186 | — |
| 12 | 新 | gpt-5.4-medium (codex-harness)闭源OpenAI | 1,436 | ±18 | 1,271 | — |
| 13 | ▼ 2 | minimax-m2.5开源MiniMax | 1,419 | ±10 | 4,846 | — |
| 14 | ▼ 2 | kimi-k2.5-instant开源Moonshot AI | 1,414 | ±11 | 3,547 | — |
| 15 | ▼ 2 | gpt-5.3-codex (codex-harness)闭源OpenAI | 1,406 | ±12 | 2,817 | — |
| 16 | ▼ 2 | minimax-m2.1-preview开源MiniMax | 1,401 | ±8 | 9,830 | — |
| 17 | ▼ 2 | gpt-5.2闭源OpenAI | 1,395 | ±15 | 1,644 | — |
| 18 | ▼ 1 | gpt-5-medium闭源OpenAI | 1,393 | ±12 | 3,948 | — |
| 19 | ▼ 3 | qwen3.5-397b-a17b开源Alibaba | 1,392 | ±10 | 3,692 | — |
| 20 | ▼ 2 | claude-sonnet-4-5-20250929-thinking-32k闭源Anthropic | 1,391 | ±7 | 15,434 | — |
| 21 | ▼ 1 | gpt-5.1-medium闭源OpenAI | 1,389 | ±9 | 6,468 | — |
| 22 | ▼ 3 | claude-opus-4-1-20250805闭源Anthropic | 1,388 | ±9 | 9,020 | — |
| 23 | ▼ 2 | claude-sonnet-4-5-20250929闭源Anthropic | 1,386 | ±7 | 17,248 | — |
| 24 | ▼ 2 | qwen3.5-122b-a10b开源Alibaba | 1,379 | ±13 | 2,251 | — |
| 25 | ▼ 1 | deepseek-v3.2-thinking开源DeepSeek · 另有 1 个档位 | 1,373 | ±8 | 7,014 | — |
| 26 | ▼ 3 | qwen3.5-27b开源Alibaba | 1,367 | ±14 | 2,105 | — |
| 27 | ▼ 2 | glm-4.6开源Z.ai | 1,357 | ±9 | 8,785 | — |
| 28 | ▼ 2 | gpt-5.1闭源OpenAI | 1,343 | ±7 | 13,323 | — |
| 29 | ▼ 1 | gpt-5.2-codex闭源OpenAI | 1,342 | ±8 | 7,022 | — |
| 30 | ▼ 3 | mimo-v2-flash (non-thinking)开源Xiaomi · 另有 1 个档位 | 1,341 | ±8 | 6,946 | — |
| 31 | ▼ 2 | kimi-k2-thinking-turbo开源Moonshot AI | 1,330 | ±7 | 13,909 | — |
| 32 | ▼ 2 | gpt-5.1-codex闭源OpenAI | 1,329 | ±9 | 6,553 | — |
| 33 | ▼ 2 | minimax-m2开源MiniMax | 1,312 | ±9 | 8,893 | — |
| 34 | ▼ 2 | claude-haiku-4-5-20251001闭源Anthropic | 1,309 | ±7 | 15,200 | $2.0 |
| 35 | ▼ 2 | deepseek-v3.2-exp开源DeepSeek | 1,287 | ±10 | 5,155 | — |
| 36 | ▼ 2 | qwen3-coder-480b-a35b-instruct开源Alibaba | 1,284 | ±7 | 14,903 | — |
| 37 | ▼ 2 | KAT-Coder-Pro-V1闭源KwaiKAT | 1,260 | ±15 | 1,964 | — |
| 38 | ▼ 2 | qwen3.5-35b-a3b开源Alibaba | 1,256 | ±16 | 1,730 | — |
| 39 | ▼ 2 | gemini-3.1-flash-lite-preview闭源Google | 1,254 | ±17 | 1,397 | $0.56 |
| 40 | ▼ 2 | gpt-5.1-codex-mini闭源OpenAI | 1,244 | ±17 | 1,541 | — |
| 41 | ▼ 2 | qwen3.5-flash闭源Alibaba | 1,243 | ±17 | 1,538 | — |
| 42 | ▼ 2 | grok-4-1-fast-reasoning闭源xAI | 1,236 | ±9 | 7,144 | — |
| 43 | ▼ 2 | mistral-large-3开源Mistral | 1,223 | ±20 | 1,045 | — |
| 44 | ▼ 2 | gemini-2.5-pro闭源Google | 1,206 | ±13 | 3,477 | — |
| 45 | ▼ 2 | grok-4.1-thinking闭源xAI | 1,205 | ±19 | 1,282 | — |
| 46 | ▼ 2 | devstral-2开源Mistral | 1,198 | ±16 | 1,693 | — |
| 47 | ▼ 2 | grok-4-fast-reasoning闭源xAI | 1,151 | ±23 | 968 | — |
| 48 | ▼ 2 | grok-code-fast-1闭源xAI | 1,143 | ±21 | 1,021 | — |
| 49 | ▼ 2 | devstral-medium-2507闭源Mistral | 1,100 | ±22 | 1,027 | — |
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