Options on the Prompt canvas

The Prompt canvas offers a range of options to help you customize the generated responses. These settings give you more control over the output, allowing you to tailor it to your specific needs. Whether you want to adjust the model, change the style, or manage the generated content, these additional options make the process more flexible and efficient.

Model column options

The options in the Model column enable users to tailor their experience based on specific needs, providing flexibility and control over how the outputs are produced.

The following table lists the options available in the Model column:

Icon	Feature	Description
	Rearrange	Move the model column anywhere on the canvas by clicking and dragging the icon.
	Remove Model	Remove a model from comparison.
	Bookmark a model	Set a model as your preferred model for consistent output. If no model is bookmarked, you must select one when committing a prompt version.
	Model Settings	Customize output generation by adjusting model parameters like temperature, top k, top p, and max tokens. The default values are top p: 1, top k: 5, temperature: 1, and max tokens: 256.
	Play	Regenerates the output for that single column.
	Average Response Time	Displays the average time taken for the model to generate a response.
	Average Tokens	Shows the mean number of input and output tokens used during interactions with the model.
	Copy	Copy the generated output to your clipboard.
	View in JSON format	View the prompt and response in JSON format. Clicking this icon opens a separate dialog displaying the request and response in JSON.
	Re-generate	Regenerates the output for that single cell.

Top toolbar options

The options available in the top toolbar of the prompt canvas are essential for efficiently generating and managing outputs. These options help to organize and share generated data, enhancing the overall user experience in the output generation process.

The following table lists the options available on the top toolbar of the prompt canvas:

Icon	Feature	Description
	Prompt library	Access a collection of around 70 templates for prompt creation, providing users with ready-to-use options. Click the icon to browse and select a template to customize or use as a base for your prompt.
	Prompt API	Enable seamless sharing of prompts using version-specific API keys. Fetch and manage prompts through API calls without repetitive copying. Click to open a dialog with API endpoints and select your preferred format (cURL, Python, Node.js) to integrate with your applications.
	Save to prompt library	Saves the current prompt to the library for future use. Click the three dots icon and save your prompt as a template in the "My Templates" section of the prompt library.
	Draft history	Capture and restore the state of the prompt canvas at different times, including inputs and outputs. Click the three dots icon and save the prompt canvas in draft history.
	Export as CSV	Exports the entire prompt canvas (including inputs, outputs, and metadata) into a CSV file for offline analysis or sharing. Click the three dots icon and export the prompt canvas as a CSV file.
	Share	Easily share prompts within the prompt canvas allowing you to collaborate with others or reuse them for future sessions. Click the three dots icon and share the prompt with others.
	Versions	Track and manage iterations of prompts. Click to view all versions of prompts. Select a version to compare, restore, or use as a draft for further editing.
	Commit	Finalize and save the current prompt version. Click to commit the current prompt as a new version, creating a record in Versions. When you save the current version, it will be labeled (e.g, V1, V2), and you will be able to track its evolution.
	Run	Executes the prompt to generate a response or output based on the current prompt settings.