Inference

AutomationBench

An agent benchmark for completing automation workflows in reproducible task environments.

  • In index
  • Agentic
ModelsModels with a result.19
Top resultBest result on this test.54.8%DeepSeek V4.1 Flash
Top-3 spreadPoints from first to third.1.7 pts
YearYear of release.N/A

Result and price

020406080100$0.1$0.3$1$3$10$30$100RESULTOUTPUT PRICE PER 1M TOKENS · LOG SCALE

Results

19 results
#ModelResultThe score from the source.
01DeepSeek V4.1 Flash54.8%
02Atria Dawn Preview53.8%
03MiMo-V2.6-Pro53.1%
04MiMo-V2.6-Flash52.3%
05Muse Spark 1.349.4%
06GLM-5.3-Flash48.8%
07GLM-5.348.2%
08Step 5 Preview44.0%
09GPT-6 Astra41.4%
10Claude Opus 5.540.0%
11GPT-6 Sol33.2%
12Hy4 preview32.1%
13DeepSeek V4 Pro 081331.8%
14Claude Fable 5.131.4%
15Kimi K330.8%
16Gemini 3.7 Flash30.4%
17Qwen3.8 Max27.3%
18Claude Opus 526.0%
19DeepSeek V4 Flash 073125.1%