Skip to content

How do I configure the OpenAi embedding node?#

Before you begin#

Step 1 - add the node#

  • Click + to open the nodes panel
  • Search for OpenAi embedding in the node search field.
  • Click the node to add it to the pathway editor.

Step 2 - choose your credential and model#

  • Choose your credential
  • Select a model loaded from your account

Step 3 - optional settings#

Note

The chosen model determines the availability of these options.

Option Description Additional information
Base URL Override URL set in OpenAi credential

Request options#

Option Description Additional information
Batch size Maximum documents to send in each request Learn about OpenAi batches
Strip new lines Toggle on to remove new lines of text from input
Timeout Maximum in milliseconds a request can take Set to -1 for no timeout

Response options#

Option Description Additional information
Response format The response can be formatted to plain text or JSON
Sampling temperature Control randomness in responses
Max retries Maximum number of attempts to make Choose this if Top P is not selected
Top P Control diversity via nucleus sampling Choose this if Max Retries is not selected

Token options#

Option Description Additional information
Frequency penalty
Presence penalty		 Penalise the frequency of OpenAi tokens Learn about OpenAi tokens
Maximum number of tokens Maximum tokens to generate based on the model Learn about OpenAi tokens
Presence penalty Add a positive value to increase the chance of new topics in output

Next step#

Further information#