McGill NLP

weblinx-browsergym

Last released Apr 25, 2025

BrowserGym integration for the WebLINX benchmark

agent-reward-bench

Last released Jul 11, 2025

Official library for AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories

webllama

Last released Jun 25, 2024

Llama-powered agents for automatic web browsing

instruct-qa

Last released Sep 19, 2023

Empirical evaluation of retrieval-augmented instruction-following models.

weblinx

Last released Oct 1, 2024

The official weblinx library

statcan-dialogue-dataset

Last released Apr 11, 2023

The Statcan Dialogue Dataset

safearena

Last released Feb 26, 2025

SafeArena is a benchmark for agent safety