Skip to main content
Batch process all your records to store structured outputs in Pinecone. The requirements are as follows.
  • A Pinecone account. Get an account.
  • A Pinecone API key in your Pinecone account. To create an API key, do the following:
    1. Sign in to your Pinecone account and open the API keys page.
    2. Click + API key.
    3. For API key name, enter some descriptive name for the API key.
    4. Click Create key.
    5. Copy the generated API key to some secure location. You will not be able to access this API key again after you close the dialog.
  • A Pinecone serverless index in your Pinecone account. Creating a serverless index on your own is optional. An index is not required to exist in advance. When you set up the connector, at runtime, the index behavior is as follows: For the Unstructured UI and Unstructured API:
    • Your workflow must contain an embedder node, and the embedder node must specify the embedding model that Unstructured will use to generate the embeedings.
    • If an existing index name is specified, and the number of dimensions that Unstructured generates does not match the number of dimensions that is specified in the existing index’s embedding settings, the run will fail. You must change the number of dimensions in your workflow’s embedder node or your existing index’s embedding settings to match, and try the run again.
    • If an index name is not specified, Unstructured creates a new index in your Pinecone account. The new index’s name will be u<short-workflow-id>-<short-embedding-model-name>-<number-of-dimensions>.
    For Unstructured Ingest:
    • If an existing index name is specified, and Unstructured generates embeddings, but the number of dimensions that are generated does not match the existing index’s embedding settings, the run will fail. You must change your Unstructured embedding settings or your existing index’s embedding settings to match, and try the run again.
    • If an index name is not specified, Unstructured creates a new index in your Pinecone account. The new index’s name will be unstructuredautocreated.
    If you create a new index or use an existing one, Unstructured recommends that all records in the target index have a field named record_id with a string data type. Unstructured can use this field to do intelligent document overwrites. Without this field, duplicate documents might be written to the index or, in some cases, the operation could fail altogether.
    To create a serverless index on your own, do the following:
    1. Sign in to your Pinecone account and open the Create a new index page.
    2. For Enter index name, enter some descriptive name for the index.
    3. For Configuration, select the check box labelled Custom settings, or click the tile labelled Manual configuration.
      Do not click any of the other tiles, such as text-embedding-3-large. Clicking any of these other tiles will cause Pinecone to generate embeddings instead of having Unstructured generate them. If Pinecone generates embeddings instead of Unstructured, this could cause any related Unstructured workflows to fail.
    4. For Vector type, select Dense.
    5. For Dimension, enter the number of dimensions for the embeddings that Unstructured will generate.
      The number of dimensions that you enter here must match the number of dimensions for the embedding model that you use in any related Unstructured workflows or ingestion pipelines. If these numbers do not match in both places, this could cause any related Unstructured workflows or ingestion pipelines to fail.
    6. For Metric, select cosine.
    7. Leave Capacity mode set to Serverless.
    8. You can leave Cloud provider and Region set to their default values, or you can select a cloud provider and region that is closest to you, if available.
    9. Click Create index.
  • Within a Pinecone serverless index, custom namespaces are supported but are not required.
The Pinecone connector dependencies. 
pip install "unstructured-ingest[pinecone]"
You might also need to install additional dependencies, depending on your needs. Learn more. The following environment variables:
  • PINECONE_API_KEY - The Pinecone API, represented by --api-key (CLI) or api_key (Python, in the PineconeAccessConfig object).
  • PINECONE_INDEX_NAME - The Pinecone serverless index name, represented by --index-name (CLI) or index_name (Python). If no value is provided, see the beginning of this article for the behavior at run time.
  • PINECONE_NAMESPACE_NAME - The name of any custom namespace in the Pinecone index to upsert data into, represented by --namespace (CLI) or namespace (Python). If no value is provided, the default namespace of default will be used.
Now call the Unstructured CLI or Python SDK. The source connector can be any of the ones supported.
This example uses the local source connector. This example sends files to Unstructured for processing by default. To process files locally instead, see the instructions at the end of this page. 
#!/usr/bin/env bash

# Chunking and embedding are optional.

unstructured-ingest \
  local \
    --input-path $LOCAL_FILE_INPUT_DIR \
    --output-dir $LOCAL_FILE_OUTPUT_DIR \
    --strategy hi_res \
    --chunk-elements \
    --embedding-provider huggingface \
    --num-processes 2 \
    --verbose \
    --api-key $UNSTRUCTURED_API_KEY \
    --partition-endpoint $UNSTRUCTURED_API_URL \
    --additional-partition-args="{\"split_pdf_page\":\"true\", \"split_pdf_allow_failed\":\"true\", \"split_pdf_concurrency_level\": 15}" \
  pinecone \
    --api-key "$PINECONE_API_KEY" \
    --index-name "$PINECONE_INDEX_NAME" \
    --namespace "$PINECONE_NAMESPACE_NAME" \
    --batch-size 80
For the Unstructured Ingest CLI and the Unstructured Ingest Python library, you can use the --partition-by-api option (CLI) or partition_by_api (Python) parameter to specify where files are processed:
  • To do local file processing, omit --partition-by-api (CLI) or partition_by_api (Python), or explicitly specify partition_by_api=False (Python). Local file processing does not use an Unstructured API key or API URL, so you can also omit the following, if they appear:
    • --api-key $UNSTRUCTURED_API_KEY (CLI) or api_key=os.getenv("UNSTRUCTURED_API_KEY") (Python)
    • --partition-endpoint $UNSTRUCTURED_API_URL (CLI) or partition_endpoint=os.getenv("UNSTRUCTURED_API_URL") (Python)
    • The environment variables UNSTRUCTURED_API_KEY and UNSTRUCTURED_API_URL
  • To send files to the Unstructured Partition Endpoint for processing, specify --partition-by-api (CLI) or partition_by_api=True (Python). Unstructured also requires an Unstructured API key and API URL, by adding the following:
    • --api-key $UNSTRUCTURED_API_KEY (CLI) or api_key=os.getenv("UNSTRUCTURED_API_KEY") (Python)
    • --partition-endpoint $UNSTRUCTURED_API_URL (CLI) or partition_endpoint=os.getenv("UNSTRUCTURED_API_URL") (Python)
    • The environment variables UNSTRUCTURED_API_KEY and UNSTRUCTURED_API_URL, representing your API key and API URL, respectively.
    You must specify the API URL only if you are not using the default API URL for Unstructured Ingest, which applies to Starter and Team accounts.The default API URL for Unstructured Ingest is https://api.unstructuredapp.io/general/v0/general, which is the API URL for the Unstructured Partition Endpoint. However, you should always use the URL that was provided to you when your Unstructured account was created. If you do not have this URL, email Unstructured Support at support@unstructured.io.If you do not have an API key, get one now.If you are using an Enterprise account, the process for generating Unstructured API keys, and the Unstructured API URL that you use, are different. For instructions, see your Unstructured account administrator, or email Unstructured Support at support@unstructured.io.