Loading page
Loading page
AI · Scale AI alternative
Label data and collect human feedback with your team to build better datasets for AI models.
Argilla is a tool for building high-quality datasets for AI. Engineers define a dataset and the questions to ask, and domain experts answer them in a web interface. The result is labeled data and human feedback that you can use to train, tune or evaluate a model.
It covers a wide range of work: classic NLP such as text classification and named entities, LLM tasks such as RAG and preference tuning, and multimodal tasks such as text to image. Annotators can filter records, see AI suggestions, and use semantic search to find similar items.
Argilla is for machine learning teams who know that output quality starts with data quality. It suits teams where the people who know the subject, such as doctors, lawyers or support leads, are not the people who write the code.
Training data is often private: customer messages, internal documents, medical notes. With Argilla in your own service, that data never goes to a labeling vendor. The project's own message is that you should own both your data and your models.
The trade-off: the project is now in maintenance mode, with fixes but no new features, so weigh that before you build on it. This template runs four containers, and Elasticsearch is memory hungry. You look after updates and backups of the database and search index.
Containers: argilla/argilla-server:latest, docker.elastic.co/elasticsearch/elasticsearch:8.12.2, postgres:15-alpine, redis:7-alpine
Step 1
Choose the Starter plan. It costs $20 a month and runs up to 3 apps.
Step 2
Open App templates in the dashboard, choose Argilla, and click Deploy. We set up the containers, secrets and domain.
Step 3
Wait a few minutes for the first deploy, then open your new URL and finish the Argilla setup screen.