排名Rank 综合排名 / 区间Overall Rank / Range 模型Model 确认成功Confirmed Success 赞扬/抱怨Praise vs Complaint 可控性Steerability Bash 恢复Bash Recovery 工具幻觉Tool Hallucination 会话数Sessions
🥇 12
10 - 21
Hy4 preview
腾讯 · Apache 2.0
4.10%±0.87% 7.53%±1.69% 7.45%±3.13% 0.70%±1.82% 6.22%±0.51% 0.02%±0.32% 54.5K
🥈 13
10 - 18
Kimi K3 (Max)
月之暗面 · Kimi K3 license
4.03%±0.53% 9.18%±1.10% 7.35%±1.80% 0.54%±1.12% 3.83%±0.50% 0.35%±0.03% 130.6K
🥉 15
10 - 21
Deepseek V4.1 Flash (Max)
DeepSeek · MIT
3.86%±0.57% 12.08%±1.11% 2.90%±1.97% 3.29%±1.34% 7.39%±0.33% 0.22%±0.04% 90.3K
4 16
10 - 21
GLM 5.2 (Max)
智谱 ZAI · MIT
3.84%±0.69% 3.78%±1.45% 9.84%±2.40% 4.18%±1.33% 1.08%±0.69% 0.35%±0.03% 84.1K
5 19
11 - 24
GLM 5.3 (Max)
智谱 ZAI · MIT
2.78%±0.64% 5.99%±1.42% 5.69%±2.19% 2.81%±1.25% 0.93%±0.57% 0.35%±0.03% 81.7K
6 21
11 - 28
DeepSeek V4 Pro (High) (0813)
DeepSeek · MIT
1.74%±1.73% 1.28%±3.84% 3.11%±6.11% 1.76%±3.82% 4.77%±1.11% 0.35%±0.03% 7.2K
7 28
22 - 31
GLM 5.3 Flash
智谱 ZAI · MIT
0.08%±0.50% 6.64%±1.11% 0.44%±1.62% 1.06%±1.04% 5.88%±0.54% 0.35%±0.03% 76.6K
8 30
26 - 33
Qwen3.8 Flash Next
阿里巴巴 · Qwen-community-1.0
0.88%±0.65% 5.52%±1.41% 2.87%±2.21% 3.42%±1.48% 2.77%±0.42% 0.90%±0.09% 80.3K
9 31
26 - 33
DeepSeek V4 Pro
DeepSeek · MIT
1.23%±0.97% 3.71%±2.28% 1.59%±3.33% 2.82%±1.84% 2.16%±0.90% 0.18%±0.12% 35.7K
10 33
29 - 33
Qwen 3.8 27B
阿里巴巴 · Apache 2.0
1.54%±0.60% 2.44%±1.37% 2.63%±2.00% 0.97%±1.30% 6.28%±0.58% 0.28%±0.12% 55.2K
11 37
36 - 42
Hy3
腾讯 · Apache 2.0
5.75%±1.16% 11.30%±2.66% 7.34%±3.46% 1.35%±2.45% 5.91%±2.27% 2.85%±1.00% 29.1K
12 38
37 - 42
Mimo V2.5 Pro
小米 · MIT
6.61%±0.86% 11.12%±2.18% 10.56%±2.49% 4.52%±1.69% 5.90%±1.51% 0.92%±0.27% 44.6K
13 39
37 - 42
Minimax M3
MiniMax · MiniMax Community License
6.81%±0.84% 14.28%±2.28% 13.18%±2.44% 6.99%±1.71% 0.51%±0.96% 0.09%±0.19% 41.0K
14 43
42 - 45
Inkling Small
Thinky · Apache 2.0
9.60%±1.35% 21.26%±4.01% 20.52%±3.61% 11.51%±3.14% 5.72%±1.23% 0.42%±0.32% 13.1K
15 44
43 - 45
Inkling
Thinky · Apache 2.0
11.00%±0.99% 21.89%±2.79% 22.04%±2.47% 11.79%±2.20% 0.73%±1.62% 0.03%±0.13% 42.7K
16 45
43 - 45
Mistral Medium 3.5
Mistral · Modified MIT
11.95%±1.46% 16.36%±3.69% 15.47%±3.98% 7.18%±3.25% 18.26%±2.70% 2.48%±0.82% 10.9K
🔍

没有找到匹配的模型,换个关键词试试?No matching models found. Try a different keyword?