1  ---
  2  title: "JinaDocumentEmbedder"
  3  id: jinadocumentembedder
  4  slug: "/jinadocumentembedder"
  5  description: "This component computes the embeddings of a list of documents and stores the obtained vectors in the embedding field of each document. It uses Jina AI Embeddings models.  The vectors computed by this component are necessary to perform embedding retrieval on a collection of documents. At retrieval time, the vector representing the query is compared with those of the documents to find the most similar or relevant documents."
  6  ---
  7  
  8  # JinaDocumentEmbedder
  9  
 10  This component computes the embeddings of a list of documents and stores the obtained vectors in the embedding field of each document. It uses Jina AI Embeddings models.  The vectors computed by this component are necessary to perform embedding retrieval on a collection of documents. At retrieval time, the vector representing the query is compared with those of the documents to find the most similar or relevant documents.
 11  
 12  |  |  |
 13  | --- | --- |
 14  | **Most common position in a pipeline** | Before a [`DocumentWriter`](../writers/documentwriter.mdx)  in an indexing pipeline |
 15  | **Mandatory init variables** | "api_key": The Jina API key. Can be set with `JINA_API_KEY` env var. |
 16  | **Mandatory run variables** | “documents”: A list of documents |
 17  | **Output variables** | “documents”: A list of documents (enriched with embeddings)  <br /> <br />”meta”: A dictionary of metadata |
 18  | **API reference** | [Jina](/reference/integrations-jina) |
 19  | **GitHub link** | https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/jina |
 20  
 21  ## Overview
 22  
 23  `JinaDocumentEmbedder` enriches the metadata of documents with an embedding of their content. To embed a string, you should use the [`JinaTextEmbedder`](jinatextembedder.mdx). To see the list of compatible Jina Embeddings models, head to Jina AI’s [website](https://jina.ai/embeddings/). The default model for `JinaDocumentEmbedder` is `jina-embeddings-v2-base-en`.
 24  
 25  To start using this integration with Haystack, install the package with:
 26  
 27  ```shell
 28  pip install jina-haystack
 29  ```
 30  
 31  The component uses a `JINA_API_KEY` environment variable by default. Otherwise, you can pass an API key at initialization with `api_key`:
 32  
 33  ```python
 34  embedder = JinaDocumentEmbedder(api_key=Secret.from_token("<your-api-key>"))
 35  ```
 36  
 37  To get a Jina Embeddings API key, head to https://jina.ai/embeddings/.
 38  
 39  ### Embedding Metadata
 40  
 41  Text documents often come with a set of metadata. If they are distinctive and semantically meaningful, you can embed them along with the text of the document to improve retrieval.
 42  
 43  You can do this easily by using the Document Embedder:
 44  
 45  ```python
 46  from haystack import Document
 47  from haystack_integrations.components.embedders.jina import JinaDocumentEmbedder
 48  
 49  doc = Document(content="some text", meta={"title": "relevant title", "page number": 18})
 50  
 51  embedder = JinaDocumentEmbedder(
 52      api_key=Secret.from_token("<your-api-key>"),
 53      meta_fields_to_embed=["title"],
 54  )
 55  
 56  docs_w_embeddings = embedder.run(documents=[doc])["documents"]
 57  ```
 58  
 59  ## Usage
 60  
 61  ### On its own
 62  
 63  Here is how you can use the component on its own:
 64  
 65  ```python
 66  from haystack_integrations.components.embedders.jina import JinaDocumentEmbedder
 67  
 68  doc = Document(content="I love pizza!")
 69  
 70  document_embedder = JinaDocumentEmbedder(api_key=Secret.from_token("<your-api-key>"))
 71  
 72  result = document_embedder.run([doc])
 73  print(result["documents"][0].embedding)
 74  
 75  ## [0.017020374536514282, -0.023255806416273117, ...]
 76  ```
 77  
 78  :::note
 79  We recommend setting JINA_API_KEY as an environment variable instead of setting it as a parameter.
 80  
 81  :::
 82  
 83  ### In a pipeline
 84  
 85  ```python
 86  from haystack import Pipeline
 87  from haystack.document_stores.in_memory import InMemoryDocumentStore
 88  from haystack_integrations.components.embedders.jina import JinaDocumentEmbedder
 89  from haystack_integrations.components.embedders.jina import JinaTextEmbedder
 90  from haystack.components.writers import DocumentWriter
 91  from haystack.components.retrievers.in_memory import InMemoryEmbeddingRetriever
 92  
 93  document_store = InMemoryDocumentStore(embedding_similarity_function="cosine")
 94  
 95  documents = [
 96      Document(content="My name is Wolfgang and I live in Berlin"),
 97      Document(content="I saw a black horse running"),
 98      Document(content="Germany has many big cities"),
 99  ]
100  
101  indexing_pipeline = Pipeline()
102  indexing_pipeline.add_component(
103      "embedder",
104      JinaDocumentEmbedder(api_key=Secret.from_token("<your-api-key>")),
105  )
106  indexing_pipeline.add_component("writer", DocumentWriter(document_store=document_store))
107  indexing_pipeline.connect("embedder", "writer")
108  
109  indexing_pipeline.run({"embedder": {"documents": documents}})
110  
111  query_pipeline = Pipeline()
112  query_pipeline.add_component(
113      "text_embedder",
114      JinaTextEmbedder(api_key=Secret.from_token("<your-api-key>")),
115  )
116  query_pipeline.add_component(
117      "retriever",
118      InMemoryEmbeddingRetriever(document_store=document_store),
119  )
120  query_pipeline.connect("text_embedder.embedding", "retriever.query_embedding")
121  
122  query = "Who lives in Berlin?"
123  
124  result = query_pipeline.run({"text_embedder": {"text": query}})
125  
126  print(result["retriever"]["documents"][0])
127  
128  ## Document(id=..., mimetype: 'text/plain',
129  ##  text: 'My name is Wolfgang and I live in Berlin')
130  ```
131  
132  ## Additional References
133  
134  🧑‍🍳 Cookbook: [Using the Jina-embeddings-v2-base-en model in a Haystack RAG pipeline for legal document analysis](https://haystack.deepset.ai/cookbook/jina-embeddings-v2-legal-analysis-rag)