Text generation allows you to create prompts that output natural language responses. Use this for chatbots, content generation, or any scenario where the output is text.Example:
Copy
---name: text-exampletext_config: model_name: gpt-4o-mini---<User>What is the capital of France?</User>
Object generation is used when you want the output to be structured data, such as JSON objects. This is useful for API responses, data extraction, or structured workflows.Example:
Copy
---name: object-exampleobject_config: model_name: gpt-4o-mini schema: type: object properties: city: type: string country: type: string---<User>Extract the city and country from: "Paris, France"</User>
Speech generation allows you to create prompts that output audio responses, converting text to speech. This is useful for voice assistants, accessibility features, or any scenario where spoken output is needed.Example:
Copy
---name: speech-examplespeech_config: model_name: tts-1-hd voice: "nova" speed: 1.0 output_format: "mp3"---<System>Please read this text aloud.</System><SpeechPrompt>This is a test for the speech prompt to be spoken aloud.</SpeechPrompt>
Below is an example of the prompt creation interface, where you can select the generation type:For more details on all generation types, see the Generation Types Overview.