Skip to main content
​
01.05.2026

​
Appended Messages for A/B Testing

The Prompt Playground now supports appending conversation history from dataset examples to your prompts, enabling powerful A/B testing workflows for models and system prompts.

​
Key Features

  • Append Conversation History: Specify a dot-notation path to messages in your dataset examples (e.g., messages or input.messages)
  • OpenAI Message Format: Full support for user, assistant, system, and tool messages
  • A/B Testing Models: Compare how different models respond to the same conversation history
  • Testing System Prompts: Evaluate different system prompts against identical user conversations

​
How to Use

  1. Load a dataset containing conversation messages in OpenAI format
  2. Click the settings button (gear icon) in the experiment toolbar
  3. Enter the path to your messages (e.g., messages)
  4. Run your experiment to see results across all prompt variants

​
Example Dataset Format

{
  "messages": [
    {"role": "user", "content": "What is the weather in San Francisco?"},
    {"role": "assistant", "content": "Let me check that for you."},
    {"role": "user", "content": "Thanks! Also, what about New York?"}
  ]
}

Using the Playground

Learn more about appending conversation history

Test a Prompt

See A/B testing examples