Embedding Models

RealTimeX supports a variety of embedding model providers out of the box with minimal setup.

Embedding models transform text into numerical vectors for storage and fast similarity search in your vector database—this is the foundation of Retrieval Augmented Generation (RAG).

Supported Embedding Model Providers

Local Embedding Model Providers

Built-in (default)

Default RealTimeX embedder, works instantly—no extra setup required.

Ollama

Modern local LLM runner supporting a range of embedding models.

LM Studio

Desktop app for local inference and embedding generation.

Local AI

Open source local inference backend for flexible, private embedding.

Cloud Embedding Model Providers

OpenAI

State-of-the-art text embedding models (e.g. text-embedding-3).

Azure OpenAI

Microsoft’s enterprise-grade OpenAI API for embedding and language.

Cohere

Powerful multilingual and domain-specific embedding models.

Google Gemini Pro

Google’s advanced multi-modal embedding models.

Hugging Face

Run any open embedding model from Hugging Face Hub.

Together AI

Multi-model cloud embedding with simple API.

OpenRouter

Unified API for embedding models across providers.

Mistral API

Fast, powerful open models for language and embedding.

Groq

Ultra-fast inference for embedding and LLM models.