综合榜单

486 个榜单 · 659 个上榜模型
1Kimi K2.61,0512Qwen3.7 Max1,0423DeepSeek V4 Pro1,0324MiMo-V2.5-Pro1,0315GLM-5.11,0276Claude Opus 4.71,0197Gemini 3.1 Pro Preview1,0138DeepSeek V4 Flash1,0119Qwen3.6 Flash1,01110GLM-5.21,009
名次 模型 机构 Elo 来源
1 Kimi K2.6 Moonshot AI 1,051 —
2 Qwen3.7 Max Alibaba Cloud / Qwen Team 1,042 —
3 DeepSeek V4 Pro DeepSeek 1,032 —
4 MiMo-V2.5-Pro Xiaomi 1,031 —
5 GLM-5.1 Zhipu AI 1,027 —
6 Claude Opus 4.7 Anthropic 1,019 —
7 Gemini 3.1 Pro Preview Google 1,013 —
8 DeepSeek V4 Flash DeepSeek 1,011 —
9 Qwen3.6 Flash Alibaba Cloud / Qwen Team 1,011 —
10 GLM-5.2 Zhipu AI 1,009 —
11 Gemini 3.5 Flash Google 1,009 —
12 Claude Opus 4.8 Anthropic 1,006 —
13 MiMo-V2.5 Xiaomi 1,004 —
14 GPT-5.5 OpenAI 1,004 —
15 Claude Fable 5 Anthropic 1,002 —
16 MiniMax-M3 MiniMax 1,001 —
17 Claude Opus 4.8 (Fast) Anthropic 1,001 —
18 Seed-2.1-Pro ByteDance 1,000 —
19 Qwen3.5 Plus Alibaba Cloud / Qwen Team 1,000 —
20 Claude Opus 4.7 (Fast) Anthropic 999 —
21 GPT-5.5 Pro OpenAI 996 —
22 MiniMax-M2.7 MiniMax 996 —
23 ERNIE 5.1 Baidu 988 —
24 Hy3 preview Tencent 986 —
25 Step 3.7 Flash StepFun 985 —
26 Seed-2.0-Pro ByteDance 971 —
27 Seed-2.0-Lite ByteDance 968 —
28 Step 3.5 Flash StepFun 967 —
29 Qwen3.6 35B-A3B Alibaba Cloud / Qwen Team 965 —
30 Grok 4.3 xAI 957 —
31 Seed-2.0-Mini ByteDance 950 —

更多基准见 全部评测基准;评测分数为事实数据,整理自公开评测体系。