LiteLLM Proxy Embedding Configuration
PipesHub can generate document embeddings through a LiteLLM Proxy instance, using the same gateway that serves your chat models. This keeps every model key and spending limit in one place.
Before you start
You need a running LiteLLM Proxy with at least one embedding model in its model_list. See LiteLLM Proxy for text generation for setup, and the LiteLLM embedding documentation for supported models.
Required Fields
Endpoint URL *
The address where your LiteLLM Proxy is reachable.
Default: http://host.docker.internal:4000
Use host.docker.internal rather than localhost when the proxy runs on the Docker host, because PipesHub itself runs in a container.
API Key *
The master key or a virtual key from your LiteLLM Proxy.
Model Name *
The embedding model’s name as defined under model_name in your proxy configuration — for example text-embedding-3-small or azure-embeddings.
Optional Fields
Model Friendly Name
A label shown in the PipesHub interface so you can tell several configurations apart. If you leave it blank, the model name is used.
Dimensions
The number of dimensions in the vectors this model produces. PipesHub uses it to size the vector collection.
Changing the embedding model or its dimensions after documents are indexed means existing vectors no longer match new ones. Re-index your content after switching.
Is Multimodal
Turn this on if the model can embed images as well as text. It is off by default.
