0

Mind2Web: Towards a Generalist Agent for the Web

Active

A dataset for developing and evaluating generalist agents for the web that can follow language instructions to complete complex tasks on any website.

Open
Publisher
The Ohio State University
Domain
Assistants
License
mit
Published
Mar 2025
Notable for
Benchmark for evaluating Assistants.
Canonical
github.com/UKGovernmentBEIS/inspect_evals/tree/main/src/inspect_evals/mind2web

Cite

Notes

Only stored in your browser.

Attribution

Attribution policy →

Related tools

3
View all

Implementations, trainers, datasets and scaffolds linked to this eval.

Mind2web RL Env (Browserbase)

Browserbase

onlineMind2Web browser benchmark for web navigation tasks with difficulty levels

ImplementationRL EnvBrowserBrowserbaseMind2web

Mind2web RL Env (Community)

onlineMind2Web browser benchmark for web navigation tasks with difficulty levels

ImplementationRL EnvBrowserBrowserbaseMind2web

ANTI BOT RL Env (Browserbase)

Browserbase

onlineMind2Web browser benchmark with filtered dataset (276 tasks from sites without anti-bot protection)

Trains towardRL EnvBrowserBrowserbaseNo Anti Bot

FAQ

What is Mind2Web: Towards a Generalist Agent for the Web?
A dataset for developing and evaluating generalist agents for the web that can follow language instructions to complete complex tasks on any website.
How can a model improve its Mind2Web: Towards a Generalist Agent for the Web score?
Tools linked to Mind2Web: Towards a Generalist Agent for the Web on Sophon include Mind2web RL Env (Browserbase), Mind2web RL Env (Community), ANTI BOT RL Env (Browserbase) - RL environments, datasets, and scaffolds that target this eval.
What license is Mind2Web: Towards a Generalist Agent for the Web under?
Mind2Web: Towards a Generalist Agent for the Web is available under mit.