Skip to main content
The Playground page is a live IDE for testing your prompts without writing any code. It connects directly to the gateway, so every run is a real request that shows up in your Request Logs with accurate token counts and cost.

​
Running a Request

  1. Select a Provider from the dropdown (OpenAI, Anthropic, Google, xAI, Groq, Together AI).
  2. Choose a Model from the list of models available for that provider.
  3. Write a System Prompt (optional) in the system prompt panel.
  4. Write your User Message in the input box.
  5. Click Run — the response appears in real time on the right side.

​
Adjusting Parameters

Use the Parameters panel on the right to adjust:
  • temperature — Controls randomness (0 = deterministic, 2 = very creative)
  • max_tokens — Maximum number of tokens in the response

​
Saving to Prompts

If you’re happy with a prompt you’ve written in the Playground, click Save as Prompt to add it to your Prompts library with a name and description.