Quickstart
To access RunLLM today, you'll need to reach out to us. RunLLM will be publicly available in early 2025, and the getting stated guide will be updated then.
Creating your first RunLLM assistant is easy. All you'll need is a one-sentence description of your product and a link to your documentation. Once you've been granted access to RunLLM, you will see a welcome page that will walk you through creating your first assistant.
Step 1: Basic information
First you'll need to tell RunLLM the name of your product and give it a sentence overview of what your product does.
Step 2: Data sources
Once you've created your assistant, you'll need to add data to it. You'll see a list of all the data source types that RunLLM supports, but you'll most likely add your documentation site first. Click on the web icon, name your data sources, and enter the link to your documentation (e.g., docs.runllm.com). By default, you shouldn't need to change any other configurations, but for more detail, see the web crawler documentation.
Step 3: Data ingestion
RunLLM will redirect you to your assistant's details page and go ahead and pull in everything under the documentation URL you provided. This will take a few minutes. To see the progress RunLLM is making, you can scroll to the bottom of the details page and click on the data source you registered.
Step 4: Testing
Once your assistant has finished ingesting data, the chat button on the assistant details page will be enabled. You can click on it to try RunLLM out. You can type in a question or upload an image, and you're off to the races!
Once you've finished these steps, you can deploy RunLLM, add more data sources, or keep chatting with the assistant to test it out!