Agentic benchmark
AutomationBench (Zapier 1.0.6) leaderboard
AutomationBench, Zapier private evaluation set release 1.0.6. Every model the catalog carries a published AutomationBench (Zapier 1.0.6) value for, ranked by that value.
CategoryAgentic
MeasureTask pass rate
Tasks600+ held-out workflows across 47 simulated applications and six business domains
DifficultyLong-horizon automation
An agent benchmark for completing business workflows across simulated applications and API endpoints.
AutomationBench (Zapier 1.0.6) ranking
1 model with a published AutomationBench (Zapier 1.0.6) value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
Models ranked by their published AutomationBench, Zapier private evaluation set release 1.0.6 value| Rank | Model | Provider | Task pass rate |
|---|
| 1 |  Claude Sonnet 5.5 | Anthropic | 44.70 |
Evidence key: Observed
Rows are ordered by the value AutomationBench published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
About AutomationBench (Zapier 1.0.6)
What the benchmark measures and how its value is published, from the source’s own record.
What is AutomationBench (Zapier 1.0.6)?
AutomationBench (Zapier 1.0.6) (AutomationBench, Zapier private evaluation set release 1.0.6) is an agentic benchmark published by AutomationBench. An agent benchmark for completing business workflows across simulated applications and API endpoints. Its evaluation set is described by the source as 600+ held-out workflows across 47 simulated applications and six business domains. The source rates its difficulty as long-horizon automation.
How is AutomationBench (Zapier 1.0.6) scored?
Each model's result is published by AutomationBench as task pass rate. ModelScale prints that value to 2 decimal places, exactly as published, and never converts or rescales it. This page ranks the 1 model that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
ModelScale