What it is
BrowserGym provides browser observations and actions through a common interface. Its ecosystem connects web benchmarks with tools for creating, running and analyzing agents.
A shared interface reduces the work needed to compare agents across web tasks. Each connected benchmark still supplies its own tasks and evaluation.
How an agent works in it
Sees
Browser screenshots, structured page observations and task goals.
Does
Browser actions exposed through the common environment interface.
Feedback
Defined by the connected benchmark, not one universal BrowserGym reward.
Example task (illustrative): Run the same browser agent against several supported task suites.
Setup
Install BrowserGym and the dependencies for the benchmark you select. Some benchmarks require separately hosted websites or additional assets.
Use it with BoltzLabs
BoltzLabs does not host this project. To record an agent working on one of its tasks:
- Prepare one task from the project: its starter files, instructions, dependencies and check command.
- In Evals, choose New eval, upload the starter folder or enter its repository, then set the task and check command.
- Choose BoltzLabs models or your own endpoint. Each run gets its own machine.
- When the run finishes, open it to read the agent’s trace, file changes and check result.
Projects that need a desktop or phone, self-hosted websites, a game or a simulator need that runtime set up separately; an eval machine does not include it. Machine time uses account credits; hosted model calls are billed separately.