AutomationBench-AA 1.0.6
Agents
Artificial Analysis' AutomationBench 1.0.6 series, identified by automation_bench_1_0_6 in its canonical data schema. The 657-task implementation reports partial objective completion with a zero score for guardrail violations, distinct from earlier source revisions.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5.5 | 69.5 |
| DeepSeek V4.1 Flash | 68.9 |
| GPT-6 Astra | 68.5 |
| Grok 4.6 | 67.0 |
| Grok 4.7 | 65.6 |
| GLM-5.3 | 62.2 |
| Gemini 3.7 Flash | 62.0 |
| GPT-6 Sol | 61.7 |
| Gemini 3.8 Flash | 60.9 |
| GLM-5.3 Flash | 60.4 |
| GPT-5.6 Sol | 60.1 |
| GPT-5.6 Terra | 59.6 |
| Claude Fable 5.1 | 59.4 |
| MiMo-V2.6-Pro | 58.6 |
| Kimi K3 | 58.3 |
| Grok 4.5 | 57.9 |
| Muse Spark 1.3 | 57.9 |
| Qwen3.8 2.4T A95B | 57.2 |
| DeepSeek V4 Pro 0813 | 56.7 |
| Claude Opus 5 | 56.6 |
| DeepSeek-V4-Pro | 56.3 |
| Qwen3.8 Max (0902) | 56.2 |
| Qwen3.8-Flash-Next | 55.9 |
| Claude Fable 5 | 54.1 |
| DeepSeek-V4-Flash-0731 | 54.0 |
| GPT-6 Luna | 53.2 |
| Gemini 3.6 Flash | 53.0 |
| Step 5 Preview | 51.0 |
| Agnes 3.0 Flash (hosted) | 50.7 |
| GPT-5.6 Luna | 50.2 |
| Qwen3.8-Max | 49.2 |
| Qwen3.8 27B | 48.2 |
| DeepSeek V4 Flash Vision Exp | 47.5 |
| GPT-5.5 | 47.3 |
| Opus 4.8 | 45.6 |
| Gemini 3.5 Flash | 42.1 |
| Muse Spark 1.2 | 40.6 |
| Muse Spark 1.1 | 38.8 |
| Agnes 2.5 Pro Alpha | 38.4 |
| K2 Horizon 375B A23B | 37.2 |