| 排名Rank | 综合排名 / 区间Overall Rank / Range | 厂商Lab | 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 | |
|---|---|---|---|---|---|---|---|---|---|
| 🥇 | 1 1 - 9 |
Anthropic
Top Model: Claude Fable 5 (High)
|
12.04%±2.54% | 10.77%±4.82% | 25.38%±8.93% | 9.17%±5.13% | 13.72%±3.63% | 1.18%±0.17% | 24.4K |
| 🥈 | 4 1 - 10 |
OpenAI
Top Model: GPT 5.6 Sol (xHigh)
|
10.72%±1.79% | 9.82%±3.65% | 23.28%±6.64% | 8.88%±3.65% | 10.48%±1.24% | 1.18%±0.17% | 18.1K |
| 🥉 | 5 1 - 9 |
月之暗面
Top Model: Kimi K3 (Max)
|
10.43%±1.04% | 15.32%±2.05% | 20.17%±3.75% | 7.54%±1.90% | 7.92%±0.82% | 1.18%±0.17% | 28.2K |
| 4 | 13 8 - 17 |
智谱 ZAI
Top Model: GLM 5.2 (Max)
|
6.67%±0.89% | 8.43%±1.87% | 11.78%±3.15% | 6.11%±1.57% | 5.85%±1.10% | 1.18%±0.17% | 52.8K |
| 5 | 15 8 - 20 |
SpaceXAI
Top Model: Grok 4.5
|
5.73%±1.20% | 5.92%±2.70% | 4.37%±4.24% | 6.43%±2.33% | 10.78%±1.25% | 1.18%±0.17% | 29.5K |
| 6 | 18 15 - 22 |
DeepSeek
Top Model: Deepseek V4 Flash (High) (20260731)
|
3.93%±0.83% | 9.14%±1.96% | 2.39%±2.74% | 2.83%±1.63% | 4.14%±0.71% | 1.17%±0.17% | 33.1K |
| 7 | 22 20 - 27 |
Meta
Top Model: Muse Spark 1.1
|
1.09%±0.60% | 6.72%±1.40% | 4.99%±1.78% | 3.16%±1.14% | 5.74%±1.19% | 1.14%±0.17% | 70.8K |
| 8 | 25 21 - 31 |
阿里巴巴
Top Model: Qwen3.7 Max
|
0.01%±0.85% | 0.15%±2.10% | 4.90%±2.63% | 0.20%±1.51% | 4.67%±1.38% | 0.53%±0.26% | 30.1K |
| 9 | 27 22 - 31 |
Google
Top Model: Gemini 3.5 Flash (High)
|
0.43%±0.62% | 0.49%±1.47% | 0.35%±1.93% | 0.68%±1.11% | 1.89%±0.99% | 0.28%±0.21% | 93.4K |
| 10 | 30 23 - 36 |
腾讯
Top Model: Hy3
|
1.30%±1.28% | 2.41%±2.82% | 2.27%±4.43% | 8.52%±2.39% | 3.49%±1.61% | 1.32%±0.72% | 17.7K |
| 11 | 32 29 - 36 |
小米
Top Model: Mimo V2.5 Pro
|
2.21%±0.88% | 3.22%±2.19% | 7.10%±2.72% | 2.37%±1.62% | 1.70%±1.38% | 0.07%±0.34% | 30.8K |
| 12 | 34 29 - 36 |
MiniMax
Top Model: Minimax M3
|
2.52%±0.84% | 5.76%±2.19% | 8.17%±2.60% | 4.81%±1.59% | 5.52%±0.83% | 0.64%±0.32% | 30.5K |
| 13 | 37 37 - 39 |
Thinky
Top Model: Inkling
|
6.71%±0.91% | 12.50%±2.56% | 16.18%±2.61% | 11.81%±1.82% | 6.39%±1.06% | 0.52%±0.27% | 35.3K |
| 14 | 38 36 - 42 |
Mistral
Top Model: Mistral Medium 3.5
|
6.93%±2.00% | 9.41%±4.64% | 9.83%±5.76% | 11.23%±4.08% | 1.23%±3.23% | 2.95%±2.22% | 5.6K |
| 15 | 44 41 - 46 |
Upstage
Top Model: Solar Pro 4
|
12.10%±2.15% | 7.28%±5.74% | 23.39%±5.20% | 13.59%±4.08% | 16.50%±5.38% | 0.26%±0.66% | 4.2K |
| 16 | 45 43 - 47 |
NVIDIA
Top Model: Nemotron 3 Ultra
|
14.56%±2.51% | 15.84%±5.39% | 14.08%±6.73% | 19.68%±4.72% | 23.68%±7.02% | 0.47%±0.35% | 11.9K |
没有找到匹配的模型,换个关键词试试?No matching models found. Try a different keyword?
Anthropic
OpenAI
月之暗面
智谱 ZAI
SpaceXAI
DeepSeek
Meta
阿里巴巴
Google
腾讯
小米
MiniMax
Mistral
NVIDIA