| 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 - 6 |
Anthropic
Top Model: Claude Opus 5 (High)
|
12.29%±1.52% | 15.54%±3.15% | 19.50%±5.50% | 10.68%±2.98% | 14.58%±0.81% | 1.14%±0.18% | 19.8K |
| 🥈 | 4 1 - 7 |
月之暗面
Top Model: Kimi K3 (Max)
|
10.42%±1.03% | 15.38%±2.03% | 19.73%±3.72% | 7.50%±1.88% | 8.24%±0.85% | 1.24%±0.18% | 28.4K |
| 🥉 | 5 1 - 11 |
OpenAI
Top Model: GPT 5.6 Sol (xHigh)
|
9.97%±1.39% | 10.54%±2.84% | 22.07%±5.11% | 6.14%±2.91% | 9.84%±1.04% | 1.24%±0.18% | 26.2K |
| 4 | 13 7 - 18 |
阿里巴巴
Top Model: Qwen3.8 Max
|
6.51%±1.40% | 11.24%±3.15% | 7.26%±4.93% | 5.04%±2.73% | 8.74%±1.17% | 0.24%±0.35% | 11.2K |
| 5 | 15 8 - 18 |
SpaceXAI
Top Model: Grok 4.5
|
6.19%±1.16% | 5.58%±2.57% | 5.83%±4.12% | 6.40%±2.17% | 11.91%±1.15% | 1.24%±0.18% | 30.7K |
| 6 | 16 9 - 18 |
智谱 ZAI
Top Model: GLM 5.2 (Max)
|
6.09%±0.85% | 7.42%±1.84% | 9.82%±2.96% | 5.63%±1.51% | 6.34%±1.13% | 1.24%±0.18% | 54.4K |
| 7 | 19 16 - 23 |
DeepSeek
Top Model: Deepseek V4 Flash (High) (20260731)
|
4.23%±0.79% | 8.75%±1.90% | 1.92%±2.67% | 3.73%±1.51% | 5.53%±0.65% | 1.24%±0.18% | 41.0K |
| 8 | 20 17 - 24 |
Google
Top Model: Gemini 3.7 Flash (High)
|
3.33%±1.06% | 9.71%±2.55% | 0.22%±3.44% | 2.53%±2.11% | 3.45%±1.19% | 1.22%±0.18% | 16.9K |
| 9 | 24 23 - 28 |
Meta
Top Model: Muse Spark 1.1
|
0.99%±0.61% | 6.42%±1.43% | 5.48%±1.81% | 3.54%±1.14% | 6.35%±1.13% | 1.20%±0.18% | 75.4K |
| 10 | 32 26 - 37 |
腾讯
Top Model: Hy3
|
1.95%±1.25% | 1.90%±2.86% | 1.37%±4.17% | 8.60%±2.37% | 3.44%±1.72% | 1.32%±0.66% | 19.4K |
| 11 | 33 28 - 37 |
小米
Top Model: Mimo V2.5 Pro
|
2.10%±0.88% | 2.94%±2.17% | 7.67%±2.67% | 1.84%±1.61% | 1.85%±1.47% | 0.08%±0.33% | 32.4K |
| 12 | 36 31 - 37 |
MiniMax
Top Model: Minimax M3
|
3.08%±0.83% | 6.35%±2.20% | 9.15%±2.52% | 6.70%±1.55% | 6.09%±0.84% | 0.72%±0.31% | 32.0K |
| 13 | 38 38 - 40 |
Thinky
Top Model: Inkling
|
6.77%±0.92% | 11.27%±2.62% | 17.61%±2.61% | 13.18%±1.93% | 7.55%±1.06% | 0.65%±0.24% | 36.5K |
| 14 | 40 38 - 45 |
Mistral
Top Model: Mistral Medium 3.5
|
7.70%±1.88% | 11.68%±4.55% | 11.50%±5.36% | 11.81%±3.86% | 1.08%±3.04% | 2.41%±1.81% | 5.9K |
| 15 | 44 39 - 47 |
Upstage
Top Model: Solar Pro 4
|
10.67%±2.32% | 6.00%±5.64% | 17.84%±6.21% | 15.28%±4.65% | 13.16%±4.52% | 1.06%±3.19% | 5.1K |
| 16 | 47 46 - 49 |
NVIDIA
Top Model: Nemotron 3 Ultra
|
15.22%±2.34% | 15.43%±4.98% | 15.05%±6.12% | 20.56%±4.36% | 25.40%±6.82% | 0.32%±0.43% | 12.1K |
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
月之暗面
OpenAI
阿里巴巴
SpaceXAI
智谱 ZAI
DeepSeek
Google
Meta
腾讯
小米
MiniMax
Mistral
NVIDIA