Skip to main content
To begin testing your model, navigate to the ‘Playground’ tab, and click ‘Add Model’. Choose the correct model for testing. Screen Shot2025 11 17at12 39 28PM Pn

​
First Interaction

Type your prompt into the text input field at the bottom, then press Enter or click the ‘Submit’ button. The AI will process your request and display the results. Review the response and refine your prompt as needed. Screen Shot2025 11 17at12 49 41PM Pn

​
Response Metadata

Screen Shot2025 11 17at1 16 44PM Pn Each AI response includes:
  • Generation timestamp
  • Action buttons (copy, regenerate, share your insight)
Dev Mode includes additional metadata. Please refer to the Dev Mode Performance Metrics section for more details.

​
Interface Layout

​
Action Buttons (Top Right)

Screen Shot2025 11 17at1 13 31PM Pn
  • Clear All Chats - Removes all conversation history from current session
  • View Graph - Opens visual representation of model architecture or workflow
  • Add Model - Dropdown menu to add or switch between AI models

​
Operating Modes

The Playground supports three primary operating modes, selectable via tabs in the input area:

​
Web Mode

Purpose: Internet-connected AI interactions Use Cases:
  • Real-time information retrieval
  • Web research and fact-checking
  • Current events and news
  • API integrations requiring live data
Example: Screen Shot2025 11 17at1 36 43PM Pn

​
Dev Mode

Purpose: Development-focused interactions Use Cases:
  • Code generation and debugging
  • API documentation and examples
  • Technical problem-solving
  • Architecture discussions
  • Development best practices
Example: Screen Shot2025 11 17at1 47 33PM Pn Dev Mode Performance Metrics Screen Shot2025 11 17at1 51 38PM Pn
  • Response Time: Total time from submitting a request to receiving a full response, including search, retrieval, and generation.
  • Speed: The rate at which tokens are generated during the response, measured in tokens per second.
  • Search Time: Time spent retrieving information from the knowledge base or context sources.
  • Strategy: The retrieval method used for the query, such as Graph RAG, Vector Search, Hybrid, or Direct.
  • Context Chunks: The number of retrieved knowledge segments used to generate the response.
  • Total Tokens: The combined count of input and output tokens used in the request.

​
Image Mode

Purpose: AI image generation Use Cases:
  • Creating visual mockups
  • Generating diagrams and illustrations
  • Producing marketing materials
  • Visualizing concepts and ideas
Downloading Generated Images
  1. Locate the ‘Download’ button below the image
  2. Click to save the image to your local system
  3. Image saves with the displayed filename
Example: Screen Shot2025 11 17at1 29 20PM Pn

​
Core Features

​
Multi-Turn Conversations

The system maintains context across multiple interactions within a session. How to use:
  • Continue asking follow-up questions without repeating context
  • Reference previous responses using terms like “that”, “the previous example”
  • Build iteratively on earlier responses
Example Flow: 
User: "Explain  OAuth 2.0  framework"
AI: [Provides explanation]
User: "Now show me how to incorporate OAuth 2.0 in my API workflow"
AI: [Provides OAuth 2.0 example with context]