Run Prompts

After setting up your prompt in the Editor, click on Run in the Playground to see the LLM’s response: While the prompt is running, you can interrupt at in any moment by clicking the Stop generation button: This immediately terminates the current execution, without saving any partial results. The system returns to its state before the execution began. This is particularly useful when a prompt is taking too long to process or when you notice an error in your setup.

​

Create a Conversation Thread

You can create a conversation thread with the LLM in the Playground to refine the output response. The process is the same as you can perform in the Editor. Learn more about how to do so in the creating role-based prompts section.

​

Prompts Execution Options

The Playground provides you with the following prompts’ execution options: Run: Use this when running a new prompt or when you want to continue the conversation, after making modifications. Clear & run: Use this when you want a fresh start. This option:

• Clears all the previous interactions and the conversation history.
• Resets the context to its initial state.
• Runs your prompt as if it is the first interaction.
• Removes any accumulated context that might influence the LLM’s response.

​

Playground History

The runs history shows all your playground runs in chronological order, with the newest runs at the top. This makes it easy to find your recent work first and even restore it. The history is saved automatically, so you do not need to worry about manually tracking your progress. To access your Playground history, click on History: The history panel will open, displaying a list of all your past runs:

​

Restoration Options

You can restore to a previous version of the history. When clicking on Restore, the system will restore both the Editor and the Playground to the selected history version. Choose among the following restoration options:

• Pretty: Visualize the differences between the last prompt and the one you want to restore as messages.
• JSON: Visualize the differences between the last prompt and the one you want to restore as a JSON.
• YAML: Visualize the differences between the last prompt and the one you want to restore as a YAML.

Choosing one of those options allows you to better visualize the differences between the current prompt and the one you would like to restore. The better choice only depends on the prompt itself. Try all of them to find the best one that suits your specific case.