Node ReferenceVectorization
Embedder Nodes
Document and text embedder nodes for every supported provider, with node types and required connections.
Embedders convert content into vector embeddings. Document embedders take documents (list[Document]) and return the same documents with embeddings attached — used in indexing workflows. Text embedders take a query (string) and return an embedding (list[float]) — used in retrieval workflows ahead of a retriever node.
| Provider | Document embedder | Text embedder | Connection type |
|---|---|---|---|
| OpenAI | dynamiq.nodes.embedders.OpenAIDocumentEmbedder | dynamiq.nodes.embedders.OpenAITextEmbedder | OpenAI |
| Bedrock | dynamiq.nodes.embedders.BedrockDocumentEmbedder | dynamiq.nodes.embedders.BedrockTextEmbedder | AWS |
| Cohere | dynamiq.nodes.embedders.CohereDocumentEmbedder | dynamiq.nodes.embedders.CohereTextEmbedder | Cohere |
| Hugging Face | dynamiq.nodes.embedders.HuggingFaceDocumentEmbedder | dynamiq.nodes.embedders.HuggingFaceTextEmbedder | HuggingFace |
| Mistral | dynamiq.nodes.embedders.MistralDocumentEmbedder | dynamiq.nodes.embedders.MistralTextEmbedder | Mistral |
| IBM watsonx | dynamiq.nodes.embedders.WatsonXDocumentEmbedder | dynamiq.nodes.embedders.WatsonXTextEmbedder | WatsonX |
| Gemini | dynamiq.nodes.embedders.GeminiDocumentEmbedder | dynamiq.nodes.embedders.GeminiTextEmbedder | Gemini |
| VertexAI | dynamiq.nodes.embedders.VertexAIDocumentEmbedder | dynamiq.nodes.embedders.VertexAITextEmbedder | VertexAI |
For where embedders sit in a RAG pipeline, see How nodes connect and Chunking and embedding.