Embedder Setup

This section explains how to connect embedding providers in RealTimeX.

Use it when you need to:

choose the default system embedder
connect a cloud or local embedding backend
select the default embedding model for that provider
find the provider-specific setup page for the backend you want to use

Open the main selector from Settings > AI Providers > Embedder.

💡

Use Embedding Models when you want the behavior guide for retrieval impact, vectorization strategy, and re-embedding consequences. This section is for connection and setup.

What this setting controls

The Embedder page is the instance-wide source of truth for how RealTimeX vectorizes documents and workspace content.

From that page, you configure:

the system embedding provider
provider credentials, endpoints, or base URLs
the default embedding model used by that provider
chunk-length limits exposed by some providers
plugin-provided embedding providers when plugins register them

Embedding settings are system-wide. Unlike LLMs, they are not normally overridden per workspace.

Common setup flow

Most embedding providers follow the same setup pattern:

Open Settings > AI Providers > Embedder.
Select the provider you want to use.
Enter the required connection details.
Choose the embedding model, or type the model name when the provider expects manual entry.
Save the configuration.
Re-embed content when changing away from an existing provider or model.

Depending on the provider, RealTimeX may load the model list only after you enter a valid API key, endpoint, or base URL.

Important warning before switching

⚠️

Changing the embedding provider or embedding model after documents are already embedded is disruptive. RealTimeX may require you to clear and re-embed stored content so retrieval stays consistent.

Local setup guides

Built-in (default)

Private, high-performance embedding engine—no setup needed.

Ollama

Run local, open-source embedding models.

LM Studio

Easy desktop app for running embedding models locally.

Local AI

Flexible, open-source backend for local embedding inference.

llama.cpp

Use a running llama.cpp server and one of its loaded embedding models.

Cloud setup guides

OpenAI

Industry-standard cloud embeddings (Ada, Babbage, etc).

Azure OpenAI

Enterprise-grade embedding via Azure’s managed OpenAI service.

Cohere

Advanced multilingual and domain-specific embeddings.

Additional providers you may also see in the app

The current in-app selector can include additional providers that do not all have dedicated setup guides in this section yet, including:

Gemini
Voyage AI
Mistral AI
LiteLLM
Generic OpenAI

Those still follow the same core pattern:

choose the provider
enter the required API key or endpoint
select or enter a model
save and re-embed when appropriate