ActBench C
AI benchmark
- Built by
- City University of Hong Kong Zhejiang University
- Tier
- C why
- Snapshot
What is ActBench?
ActBench measures behavioral safety of cowork agents by scoring execution trajectories on matched benign and adversarial tasks for risks like data disclosure, unauthorized state changes and unauthorized API calls.
Security mostly
“Cowork agents may complete benign tasks while disclosing protected data, manipulating unauthorized state, invocate unauthorized API.”
arxiv.org/abs/2608.09476v1 ↗General capability
“ActBench is a self-evolving benchmark of **behavioral safety in cowork agents**.”
github.com/zjuicsr/ActBench ↗
Who built ActBench
Led by Zhejiang University and City University of Hong Kong.
Paper: arxiv.org/abs/2608.09476v1 ↗
Tier C
Niche: a handful of models scored, or built by an RL vendor.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- 12 models scored
- in curated RL directories (1 pt)
How we got to tier C
- Adoption: 12 models scored. Tier C.
- Not yet read on X.
- In curated RL directories: 1 point, a floor of none.
- Confirmed: a line from its own site or paper places it.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Where ActBench sits
Its domain on the map, live. Hover a logo for its name; click to open it. Open it on the full map
Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-10-19 snapshot.