Skip to main content

Enhanced OpenAI channel

Beta

This is a beta feature, which means we’re still developing it. Some functionality might change. 

To get access, register at quickbase.com/beta. After you register, we’ll contact you if you’re a match for the program. Contact your account team to learn more.

Use the Enhanced OpenAI channel to integrate with Open AI to unlock a wide range of benefits, including powerful language understanding, automation, and decision-support capabilities.

In this article

How to connect

  1. On the My pipelines page, select Create Pipeline.
  2. Select the OpenAI-Beta channel from your list of channels.
  3. Select Invoke an operation.
    invoke-an-operation.png
  4. In the Connection section of the step, enter your API key and select Connect. If you have a private OpenAI instance, you can also enter the Instance URL.
    openai-connection.png
  5. In the Account field, ensure the account you just connected is selected.

Congrats, you've successfully connected your channel and you can start using it.

Tip

If any of the steps fail, try again. The channel may connect immediately, or you might need to repeat the steps. We apologize for any inconvenience and thank you for your understanding as we work to improve the user experience during beta.

How to reconnect

You may need to reconnect your account to a channel. Reasons may be (but not limited to):

  • If you need to connect a different account.
  • Authorization updates, such as a changed password.
  • Editing the access rights that Pipelines has to the channel.

To reconnect:

  1. Select a pipeline that already has this channel in it.
  2. Open a step that contains this channel.
  3. Under account, select Connect (or reconnect) and follow the process above, How to connect.

Steps

This table lists the steps you can use with the Enhanced OpenAI beta.

Type Name Description
Action Creates a completion for the provided prompt and parameters. This action generates a completion based on the given prompt and parameters, using a model to predict and produce the most suitable continuation of the text.
Action Creates an image given a prompt. This action generates an image based on the provided prompt, using a model to interpret the prompt and create a visual representation.
Action Lists the currently available models, and provides basic information about each one such as the owner and availability. This action retrieves a list of all available models, along with essential details like the owner and their availability status.
Action Retrieves a model instance, providing basic information about the model such as the owner and permissioning. This action fetches a specific model instance and provides key information about it, including the owner and permission settings.
Action Classifies if text and/or image inputs are potentially harmful. Learn more in the [moderation guide](/docs/guides/moderation). This action assesses whether the provided text or image inputs could be harmful, using a model to classify the content. Learn more in the moderation guide.
Action Creates an invite for a user to the organization. The invite must be accepted by the user before they have access to the organization. This action generates an invitation for a user to join the organization. The user must accept the invite before gaining access to the organization.
Action Retrieves an invite. This action fetches the details of a specific invite, providing information about the invitation status and recipient.
Action Deletes an invite. If the invite has already been accepted, it cannot be deleted. This action removes an invite from the system. If the invite has already been accepted, it cannot be deleted.
Action Retrieves a user by their identifier. This action fetches the details of a user based on their unique identifier, providing information about the user's profile and status.
Action Modifies a user's role in the organization. This action changes the role of a user within the organization, updating their permissions and access levels accordingly.
Action Deletes a user from the organization. This action removes a user from the organization, revoking their access and permissions.
Action Creates an embedding vector representing the input text. This operation generates a numerical representation (embedding vector) of the input text, capturing its meaning and relationships in a way that a machine can process. It’s like converting words into a mathematical form.

Limits

  • The file size limit is 1MB
  • We currently do not provide triggers for this channel
  • The maximum OpenAI token limit is 2048

Related articles

Top