Argilla

Argilla is an open-source, Hugging Face-backed data curation and annotation tool for building high-quality NLP, LLM, and multimodal datasets with human and machine feedback.

Argilla is an open-source data-centric tool for AI makers, enabling engineers and domain experts to collaborate on building high-quality datasets. It combines human and machine feedback so teams can curate data faster and train more robust language and multimodal models. Since joining Hugging Face, Argilla shipped its 2.0 release and gained tight integration with the Hugging Face Hub, letting teams get up and running in minutes. With Argilla on Hugging Face Spaces, you can launch your own Argilla Server quickly and at no cost without local setup, and open annotation tasks to the broader community with automatic task distribution and minimum-response controls for quality. Argilla supports a wide range of AI projects: traditional NLP such as text classification and named-entity recognition, LLM workflows like RAG and preference tuning, and multimodal tasks such as text-to-image. Teams at organizations including the Red Cross, Loris.ai, and Prolific use it to improve the quality and efficiency of their AI data pipelines.