Open AI Completion

Integrate OpenAI Chat Completion into Datastreamer Pipelines

Statistics

TypeSpeedPartner Type
Post-Processing ClassifierInstantDatastreamer Internal

Example Use Cases

Compatible Data Sources

As a Post-Processing operation, it can be run on any data source containing JSON documents.

Pipeline Integration

Modify your Datastreamer Pipeline, make sure you have at least one Ingress component for Input Documents, from Pipeline Component Picker, find OpenAI Chat Completion component to connect to.

OpenAI Chat Completion Pipeline Component

Datastreamer Pipeline Component for OpenAI Chat Completion


Configure OpenAI Chat Completion Component

Fill in required fields on the component configuration panel

Supported Options

  • Text field name available in input documents and containing text content
  • Destination field name to add generated response into
  • Prompt Text
  • Model name and version
  • Max Completion Tokens
  • Top P
  • N completions
  • Log Probabilities
  • Condition, to filter input documents

For more information on parameters usage, see completion parameter. See Open AI Completion API Reference for parameters