Logo

Console

The Skytells Console is a powerful browser-based playground that allows you to interact with Skytells AI models directly from your dashboard. It provides a user-friendly interface for testing and experimenting with AI models without writing any code, featuring realtime previews and a highly dynamic interface.

Overview

The Console provides a streamlined interface for executing prompts against different Skytells AI models, allowing you to:

  • Test different prompts and parameters before integrating into your application
  • Quickly generate AI content for prototyping and experimentation
  • Compare results across different models and parameter settings
  • Understand model behavior without writing code
  • Observe generation in realtime with live progress updates
  • Interact with a responsive, dynamic interface that adapts to your inputs

Accessing the Console

  1. Log in to your Skytells Dashboard
  2. Navigate to "Console" in the left sidebar menu
  3. You'll be presented with the Console interface as shown below:

Using the Console

Selecting a Model

  1. In the "Model" dropdown, select the AI model you want to use (e.g., TrueFusion Pro)
  2. Your API key should be automatically populated

Input Methods

The Console supports multiple input methods to interact with models:

Form Input

The simplest way to interact with models. Fill in the required fields:

  1. Enter your prompt in the text area
  2. Configure additional model parameters such as aspect ratio
  3. Click "Run" to generate content

JSON Input

For more advanced configurations:

  1. Select the "JSON" tab in the input section
  2. Provide your input as a JSON object with all desired parameters
  3. Click "Run" to execute

Example JSON input for image generation:

{
  "prompt": "A futuristic cityscape with flying cars",
  "negative_prompt": "blurry, low quality, distorted",
  "steps": 30,
  "guidance": 7.5,
  "aspect_ratio": "16:9",
  "num_outputs": 1
}

Node.js / Python / HTTP Input

For testing code snippets:

  1. Select the corresponding tab (Node.js, Python, or HTTP)
  2. Write or paste your code snippet
  3. Click "Run" to execute

Advanced Options

Click the "Advanced Options" button to access additional configuration options:

Aspect Ratio

Select from predefined aspect ratios or choose a custom size for your generated images.

Image Input

For image-to-image models:

  1. Upload a reference image or provide an image URL
  2. The aspect ratio of your output will match the input image

Prompt Strength

Adjust the slider to control how strictly the model should follow your prompt:

  • Lower values (toward 0): More creative interpretation
  • Higher values (toward 1): More literal interpretation

Number of Outputs

Specify how many different outputs the model should generate for your prompt.

Realtime Feedback and Preview

The Console offers enhanced realtime capabilities that make it exceptionally powerful for iterative development:

  • Live Progress Updates: Watch as your generation progresses with status indicators
  • Preview Mode: See preliminary results as they form, even before the complete generation is finished
  • Responsive Feedback: The UI updates dynamically as processing occurs
  • Step-by-Step Visualization: For image models, observe the generation process unfold in stages

Dynamic User Interface

The Console's interface adapts intelligently to your interactions:

  • Context-Aware Controls: Options and parameters update based on the selected model
  • Interactive Elements: Sliders, toggles, and input fields provide immediate visual feedback
  • Responsive Layout: Adjusts to different screen sizes and orientations
  • Visual Indicators: Status icons and progress bars communicate current state clearly

Viewing Results

Generated content appears in the output panel on the right side:

  • For images: View and download the generated images
  • For text: View the generated text content
  • For JSON output: Examine the structured response data
  • For logs: Review processing information and any errors

The preview panel updates in realtime as content is generated, allowing you to see results as they become available rather than waiting for the entire process to complete.

Console vs. API

The Console provides the same capabilities as the Skytells API but in a user-friendly browser interface. Any successful generations in the Console can be replicated in your code using the same parameters.

Benefits of the Console:

  • No code required
  • Immediate visual feedback
  • Easy parameter experimentation
  • Learning tool for understanding model behavior
  • Realtime visualization of the generation process
  • Dynamic UI that adapts to different models and inputs

Best Practices

  1. Start Simple: Begin with basic prompts and gradually add complexity
  2. Compare Parameters: Test how different parameter values affect output quality
  3. Save Successful Configurations: Note down parameters that produce good results for use in your applications
  4. Check Logs: If something doesn't work, review the logs tab for error messages
  5. Iterate Quickly: Use the Console for rapid experimentation before implementing in code
  6. Observe Realtime Updates: Watch the generation process to understand how models progressively refine outputs
  7. Leverage the Preview: Use the preview functionality to make quick decisions about whether to continue or cancel a generation

Next Steps

Once you've found the optimal parameters for your use case:

  1. Implement the API in your application
  2. Set up webhooks for asynchronous processing
  3. Explore SDKs for your preferred programming language
  4. Review billing for information on usage and costs

Was this page helpful?