FAQ
Let's take a look at some frequently asked questions about Pipe.
What is a Pipe?
Pipe is a high-level layer to Large Language Models (LLMs) that creates a personalized AI assistant for your queries. It can leverage any LLM models, tools, and knowledge with your datasets to assist with your queries.
What is a System Prompt Instruction?
Initial setup or instruction for the LLM that configures or instructs the LLM on how to behave.
What is a User Prompt?
A text input that a user provides to an LLM to which the model responds.
What is an AI Prompt?
The LLM's generated output in response to a user prompt.
How to run Playground in Pipe?
Assuming the Pipe API keys are configured:
- Select any LLM model. By default OpenAI
gpt-4o-miniis selected. - If the Pipe is of type
generate, simply run it. - If it is a
chatpipe, write hello in Playground and run the Pipe.
Can I add readme to a pipe?
Yes, you can add readme to any Pipe.
When you create a Pipe, it already contains a readme. Go all the way down in a Pipe. You will find a readme there. Simply edit it.
Can I run experiments on a chat Pipe?
No, only generate type Pipes can run experiments.
Where can I find the Pipe API key?
Navigate to the API tab in the Pipe navbar. Here you will find Pipe API secret.
Does each Pipe have its own API key?
Yes, every Pipe you create on Langbase contains its unique API key.
Pipe Playground is not running. How can I fix it?
- Check if you have configured the LLM API key for your selected model.
- Try providing a user prompt if you are not providing it already.