Langbase

Quickstart: Build FAQs RAG using Memory

Let's build your first RAG using memory in a minute.

In this quickstart guide, you will:

  • Create a memory on Langbase.
  • Build a RAG by connecting your pipe and memory.
  • Deploy your RAG system using the Pipe API.

Let's get started

Let's create a RAG system for Pipe FAQs using memory to answer questions about Pipe.

Step #1Create a Memory

  1. Login to your account on Langbase.
  2. Click on the “Memory” tab on the left sidebar. This will open the memory page.
  3. Click on the “Add New” button. Enter a name for the new memory, let's call it pipe-faqs.
  4. Click on "Create" button to create the memory.

Step #2Add Document to Memory

  1. Go to Pipe's FAQs page and save the page as a .pdf or .txt.
  2. Open pipe-faqs memory. Click on the upload area or drag & drop documents to add them to the memory. Maximum five documents can be uploaded at a time with a maximum size of 10MB per document.
  3. Click “Refresh” to fetch latest status. Once documents are uploaded, they are processed automatically in the background. When “Ready” status appears, the document is ready to be used.
Warning

Memory requires OpenAI Key

To use the memory, you need to have an OpenAI API key. Make sure your OpenAI API key is present to your Keyset in Langbase. If not, you will be prompted to add the key when you try to add a document to the memory.

Step #3Add memory to Pipe

  1. Create a new Pipe or open an existing Pipe.
  2. Click on the "Memory" button in Pipe editor.
  3. Click on the "Search Memory Sets" dropdown and select the pipe-faqs memory from the list of memory sets.

Step #4Prompt Away!

That's it! You have successfully created a RAG system using memory. Now, you can prompt the LLM model to get answers to your questions. It will search the memory and provide you with the best possible answer to your question.

✨ Congratulations! You have successfully created a RAG system using memory. You can now deploy your RAG system using the Pipe API.

Next Steps

Dive deeper into the Memory concepts to explore more features and capabilities of memory.