Giskard

Giskard Documentation

Phare Benchmark Runner

🐢 Open-Source Evaluation & Testing library for LLM Agents

A practical tool for managing async pipelines

📜 Collection of real failure cases of LLM-based applications

The official Python SDK for the Giskard Hub API

Agent skills to streamline adoption and usage of Giskard products

Web app to run the human study of StereoTales

Pipeline to generate StereoTales dataset, find the stereotypes and evaluate harmfulness from LLMs