Pipelines transform your unstructured data to a searchable vector collection, chaining up the embedding, ingestion, search, and deletion of your data.
Understanding Pipelines [READ MORE]
Zilliz Cloud Pipelines is a robust solution for transforming unstructured data such as documents, text pieces and images into a searchable vector collection. This guide provides a detailed description of the three main Pipelines types and their functions.
Create Pipelines [READ MORE]
You can create pipelines either on the Zilliz Cloud console or via RESTful APIs. The Zilliz Cloud web UI provides a simplified and intuitive way of creating Pipelines while creating pipelines via RESTful API offers more flexibility and customization compared to the Web UI.
Manage Pipelines [READ MORE]
This guide walks you through how to manage your created pipelines.
Run Pipelines [READ MORE]
You can run the created pipelines to convert your unstructured data into embedding vectors(with or without metadata field), conduct a text semantic search, or delete unnecessary documents.
Estimate Pipelines Usage [READ MORE]
The cost of running pipelines is measured by tokens. Similar to Large Language Model (LLM) that uses token as a basic unit, pipelines process documents and search queries by parsing and embedding the text as a series of tokens. To understand the cost of a pipeline run, you can use our Estimate Pipeline Usage tool to count tokens of a file or a text string.
Pipeline Integrations [READ MORE]
Still do not know how to integrate great open-source and third-party services with Zilliz Cloud Pipelines? Use these guides to start with.