Build your first assistant in three steps
Connect a language model, build a corpus, configure the assistant and test it — the basic workflow in EasyRag.
Community
Let’s shape together how EasyRag is used
The EasyRag community is just getting started. It begins with the essentials: reliable resources to learn from, a log of platform changes, and an open channel to share your feedback and contributions.
Learn
The most useful how-to guides to get started and go further.
Preparing your documents for accurate answers
Formats, structure, duplicates, corpus priorities — the best practices that make a real difference to answer quality.
Embedding the widget on a website
The embed code to copy, its options, declaring allowed origins, and the checks to run.
Calling an assistant via the API
Create an API key, query an assistant with chat/complete, and use the sources and usage data returned.
- Display themes for the user portalThe chat portal now offers ten display themes, including an automatic mode that follows the device setting.
- vLLM support and API toolsNew vLLM provider for self-hosted models, and declarative REST API tools for agents.
- Excel export for dashboardsDashboards and reports can be exported to Excel.
Contribute
Every contribution is read and answered. Nothing is published without your consent.
Propose a guide
Solved a use case? Share your approach: we will edit it and publish it with your consent.
Report an issue
Unexpected behavior or an error in the documentation: describe it, with the version and the steps to reproduce it.
Suggest an improvement
Missing a connector, a channel or a feature? The needs you share shape our priorities.
Share your story
Your feedback helps other organizations get started. It is only published with your explicit consent.
Discussion spaces
There is no public discussion space for EasyRag yet. Rather than opening an empty forum, we are starting with direct conversations with users; a dedicated space will be offered once the community warrants it. Would you like to be notified when it opens?Let us know.
The platform’s Docker images are public: Docker Hub (opens in a new tab).
The community’s reference documentation is on the EasyRag wiki (in French) (opens in a new tab).
A community backed by its sponsors
The EasyRag community platform is published by Patrick Beaucamp. The company LaTribu-IT.fr (opens in a new tab)is its main sponsor, funding and promoting the community.
Get involved now
A question, feedback, an idea for a guide: write to us, we read everything.