WebBench B
AI benchmark · Agentic
What is WebBench?
WebBench measures how effectively browser agents complete realistic read and write web workflows across 2,454 tasks on 452 live websites.
Computer use mostly
“WebBench is an open, task-oriented benchmark designed to measure how effectively browser agents handle complex, realistic web workflows.”
github.com/Halluminate/WebBench ↗
Who built WebBench
Led by Skyvern.
Tier B
Established: 15 or more models scored, or curated RL directories rate it.
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.
- 1 models scored
- listed with Halluminate
- in curated RL directories (6 pts)
How we got to tier B
- Adoption: 1 models scored; listed with Halluminate. Tier C.
- On X: unclear, none buzz No move.
- In curated RL directories: 6 points, a floor of B. Raised to B.
- 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.
Training methods
Methods WebBench uses, offers or builds for, with public evidence: its own website says so, or two independent sites do.
- RLVR Reinforcement Learning with Verifiable Rewards On its own website
“We employ RLVR-style unit tests that deterministically evaluate the agent's final state. Complete access to the simulator's underlying data allow us to compute and check ground truth programmatically, eliminating dependence on live-si…”
halluminate.ai - GRPO Group Relative Policy Optimization On its own website
“Imitate, Then Explore: SFT and GRPO”
halluminate.ai - SFT Supervised Fine-Tuning On its own website
“Training has two phases: SFT followed by RL, specifically Group Relative Policy Optimization (GRPO).”
halluminate.ai
Where WebBench 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, 2026-11-14 snapshot.