LM Studio Embedder

️⚠️

Heads up!

LMStudio’s inference server only allows you to load multiple LLMs or a single embedding model, but not both. This means LMStudio cannot be both your LLM and embedder.

LMStudio (opens in a new tab) supports LLM and embedding GGUF models from HuggingFace that can be run on CPU or GPU.

LMStudio is a separate application that you need to download first and connect to.

Connecting to LM Studio

When running LMStudio locally, you should connect to LMStudio by first running the built-in inference server.

You must explicitly load the embedding model before starting the inference server.

You can update your model to a different model at any time in the Settings.