Benchmarks

BrowserGym

Unified interfaces for browser-agent tasks and benchmark environments.

AI-assisted editorial review against a primary source. This is a catalog annotation, not an independent reproduction or endorsement.

Original source ↗Read as Markdown Record JSON

What is this resource?

Resource type: Benchmarks
Environment: web
Topics: evaluation
Record publication date: Not recorded
Evidence retrieved: 2026-10-16T06:35:57Z
Resource ID: github:servicenow/browsergym

What licenses are recorded?

Code, training data, weights, and documents can have different terms. A blank declaration means unknown, not unrestricted use.

Code: Apache-2.0 (repository LICENSE header; not a weights/data license)
Data: Not recorded
Document: Not recorded
Weights: Not recorded

Evaluation protocol

Focus: shared browser-agent evaluation interface, benchmark adapters with separate task definitions
Setting: mixed
Observations: browser screenshot, DOM snapshot, accessibility tree, tab/URL state, previous action and its error, chat messages and task goal
Actions / outputs: element-ID actions, coordinate mouse and keyboard actions, tab/navigation/scroll operations, Python/Playwright action mode
Evaluation: Gymnasium reset/step interface delegates reward, completion, messages and task info to each adapter’s validate() method. There is no universal BrowserGym success metric or shared task total; retain benchmark/adapter version and action subset for comparisons.

Scale and counting units

Not recorded

A framework with interactive browser adapters and a static WebLINX evaluation integration, so the profile setting is mixed. README-listed integrations are not added as disjoint tasks or datasets. Core requires Playwright/Chromium; individual benchmarks have their own setup, credentials, licenses and validation rules. Apache-2.0 covers repository code only. No browser tasks or third-party project code were executed.

Compare source-reviewed profiles

Explore related topics

Evaluation

Related resources

documented by: The BrowserGym Ecosystem for Web Agent Research

Primary evidence and verification

Source retrieval is separate from experiment reproduction. Use the original publication or publisher page when citing claims about the resource.

editorial primary review
Retrieved 2026-10-16T06:35:57Z. Supports: name, kind, summary, environments, topics, scope, relations, details.benchmark_protocol.
Content fingerprint
7b685115ef822989fac9093518d16c46f0202736d005eed56e7f9fc36b43f3ee
editorial primary review
Retrieved 2026-10-17T14:22:05Z. Supports: details.benchmark_protocol.
Content fingerprint
867cc6efdfead356839fef8a26b492502b8b5e3d165d1b3c17d24aaeec268d63
editorial primary review
Retrieved 2026-10-17T14:21:11Z. Supports: details.benchmark_protocol.
Content fingerprint
c4bbe6b7d0cb36a5d5effab1aa5568301cead97fe4015273e1b82c667da3823a
editorial primary review
Retrieved 2026-10-17T14:14:08Z. Supports: details.benchmark_protocol.
Content fingerprint
418efbe8c65ad7ebab61d3a066339876c820a9c957506eb456b8078ab352f6bf
license header
Retrieved 2026-10-16T06:41:57Z. Supports: licenses.code.
Content fingerprint
b192c58991e8ff585cc574615d40e74185404d4b96c1109d423071ab1367344b

Suggest a correction

Open a source-backed correction. Include this record ID and the original source. Read our editorial policy and coverage boundaries.