HyperAI
Command Palette
Search for a command to run...
AutomationBench-AA
一项基准,衡量智能体在模拟 SaaS 应用环境中完成任务的比率,按每项任务在未违反防护规则的前提下完成的目标占比计分。
- 智能指数权重
- 5%
- 模型
- 150 当前,256 总共
- 中位数
- 46.0%
- 范围
- 0.2%–77.8%
| 排名 | 比较 | 模型 | 评分 | 成本 / 任务 | 时间 / 任务 | 输出 token / 任务 |
|---|---|---|---|---|---|---|
| 1 | Gemini 4 Carbon (High) Google | 77.8% | $1.89 | 10m 33s | 52K | |
| 2 | Gemini 4 Carbon (Medium) Google | 77.6% | $1.65 | 10m 0s | 44.3K | |
| 3 | Gemini 4 Argon (High) Google | 77.5% | $1.22 | 8m 1s | 32.6K | |
| 4 | 76.4% | $4.45 | 11m 11s | 68.9K | ||
| 5 | 74.2% | $2.27 | 5m 7s | 31.7K | ||
| 6 | Kimi K3.1 Pro Kimi | 73.3% | $0.83 | 18m 55s | 36.8K | |
| 7 | MiniMax-M3.1-Flash MiniMax | 72.5% | $0.066 | 5m 50s | 55.8K | |
| 8 | 71.8% | $1.03 | 6m 37s | 60.9K | ||
| 9 | 71.4% | $1.69 | 3m 56s | 22K | ||
| 10 | 71.3% | — | — | 19.7K | ||
| 11 | 69.5% | $1.43 | 5m 52s | 41.1K | ||
| 12 | 68.8% | $1.33 | 3m 3s | 16.6K | ||
| 13 | DeepSeek V4.1 Flash 1017 (High) DeepSeek | 68.5% | $0.14 | 6m 58s | 59.2K | |
| 14 | GPT-6 Astra (Max) OpenAI | 68.5% | $1.60 | 5m 43s | 12.9K | |
| 15 | 68.1% | $0.72 | 2m 30s | 18.3K | ||
| 16 | Muse Ember (Max) Meta | 68.1% | $0.018 | 42.8s | 19.5K | |
| 17 | 68.0% | $1.28 | 2m 51s | 15.9K | ||
| 18 | Ling 3.1 Flash 1019 InclusionAI | 67.5% | $0.15 | — | 46.7K | |
| 19 | 67.2% | $0.63 | 2m 9s | 15.6K | ||
| 20 | GPT-6 Astra (Xhigh) OpenAI | 67.2% | $1.41 | 4m 25s | 10.3K | |
| 21 | 66.7% | $0.016 | 37.9s | 17.4K | ||
| 22 | DeepSeek V4.1 Flash 1017 (Max) DeepSeek | 66.6% | $0.17 | 7m 39s | 63.4K | |
| 23 | GPT-6 Astra (High) OpenAI | 66.6% | $1.30 | 3m 54s | 9.19K | |
| 24 | GPT-6.1 Sol (Xhigh) OpenAI | 66.6% | $0.25 | 3m 10s | 10.6K | |
| 25 | Tencent Hy4 Tencent | 65.6% | $0.25 | — | 37K | |
| 26 | Grok 4.7 (Xhigh) SpaceXAI | 65.6% | $0.95 | 8m 14s | 31.6K | |
| 27 | 65.5% | $0.42 | 2m 30s | 20.8K | ||
| 28 | 65.0% | $0.88 | 3m 21s | 21.8K | ||
| 29 | GPT-6.1 Sol (Max) OpenAI | 64.9% | $0.30 | 4m 8s | 14K | |
| 30 | GPT-6 Astra (Medium) OpenAI | 64.6% | $1.18 | 3m 3s | 7.83K | |
| 31 | GPT-6.1 Sol (High) OpenAI | 64.5% | $0.23 | 2m 41s | 9.19K | |
| 32 | MiMo-V2.6-Flash Xiaomi | 64.1% | $0.017 | 11m 9s | 23.5K | |
| 33 | Grok 4.7 (High) SpaceXAI | 63.5% | $0.85 | 7m 9s | 28.8K | |
| 34 | Gemini 4 Carbon (Low) Google | 63.5% | $0.90 | 3m 34s | 13.8K | |
| 35 | 63.2% | $0.70 | 2m 38s | 16.5K | ||
| 36 | GPT-6.1 Sol (Medium) OpenAI | 62.6% | $0.20 | 2m 7s | 7.49K | |
| 37 | GLM-5.3 (Max) Z AI | 62.2% | $0.46 | 6m 44s | 27.7K | |
| 38 | Ling 3.1 Flash InclusionAI | 61.7% | — | — | — | |
| 39 | 61.2% | $0.64 | 2m 26s | 15.1K | ||
| 40 | 60.9% | $0.76 | 5m 35s | 37.6K | ||
| 41 | GLM-5.3-Flash Z AI | 60.4% | $0.032 | 6m 48s | 20.2K | |
| 42 | GPT-6.1 Terra (Max) OpenAI | 60.2% | $0.19 | 3m 44s | 16.5K | |
| 43 | Gemini 3.8 Flash (High) Google | 59.9% | $0.90 | 6m 43s | 46.3K | |
| 44 | GPT-6.1 Terra (Xhigh) OpenAI | 59.9% | $0.13 | 2m 26s | 10.4K | |
| 45 | 59.4% | $0.31 | 1m 44s | 14.7K | ||
| 46 | GPT-6 Astra (Low) OpenAI | 59.1% | $1.00 | 2m 33s | 6.08K | |
| 47 | MiMo-V2.6-Pro Xiaomi | 58.6% | $0.042 | 6m 15s | 21.1K | |
| 48 | GPT-6.1 Terra (High) OpenAI | 58.5% | $0.11 | 2m 2s | 8.52K | |
| 49 | Mistral Large 4 (High) Mistral | 57.8% | $0.18 | 3m 26s | 24.1K | |
| 50 | GPT-6.1 Terra (Medium) OpenAI | 57.8% | $0.096 | 1m 44s | 7.18K |