image-generation

Create Stunning Images with Kosmos G's Advanced Generation Capabilities

26 Apr 2025
Create Stunning Images with Kosmos G's Advanced Generation Capabilities

In the world of AI-driven creativity, Kosmos G stands out as a powerful tool for generating contextual images from multi-modal prompts. Developed by Microsoft, this multi-modal model leverages advanced perception capabilities to enable developers to create images that are not just visually appealing but also contextually relevant. The ability to generate variations and mix images opens up a realm of possibilities for applications in marketing, content creation, and even gaming. With Kosmos G, you can turn your imaginative ideas into stunning visual representations quickly and efficiently.

Prerequisites

To get started with Kosmos G, you will need an API key to access the Cognitive Actions. A basic understanding of making API calls will also be beneficial as you integrate this powerful tool into your projects.

Generate Contextual Images with Kosmos-G

The "Generate Contextual Images with Kosmos-G" action allows developers to create images based on textual prompts and existing images. This functionality is particularly useful for enhancing visual storytelling, generating marketing content, or creating unique art pieces that blend different visual elements.

Purpose: This action solves the problem of generating high-quality images that align closely with specified context, allowing for creative freedom while maintaining relevance to the intended message.

Input Requirements:

  • image1 (required): A URI of the primary input image that serves as the base for the generation.
  • image2 (optional): A URI of a secondary image that can enhance the composite.
  • prompt: A string that includes placeholders for images, guiding the generation process.
  • negativePrompt: A string for specifying elements to avoid in the generated images.
  • numberOfImages: An integer (1-4) indicating how many images to generate.
  • textGuidanceScale: A number (1-15) that influences how much the text prompt affects the final image.
  • numberOfInferenceSteps: An integer (10-100) that dictates the quality and processing time of the image generation.

Expected Output: The action will return a URI of the generated image(s), which can be used directly in applications or for further processing.

Use Cases for this specific action:

  1. Marketing Campaigns: Create visually striking images tailored to specific campaigns by combining existing brand images with creative prompts.
  2. Content Creation: Generate unique images for blog posts or social media content that align with written narratives or themes.
  3. Game Development: Use contextual image generation to create assets that are responsive to game narratives or player actions.
  4. Artistic Projects: Artists can explore new styles by mixing images and concepts, pushing the boundaries of traditional art forms.
import requests
import json

# Replace with your actual Cognitive Actions API key and endpoint
# Ensure your environment securely handles the API key
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
# This endpoint URL is hypothetical and should be documented for users
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute"

action_id = "f5959c04-6b27-4bb7-99b1-d13705733cdf" # Action ID for: Generate Contextual Images with Kosmos-G

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "seed": 20,
  "image1": "https://replicate.delivery/pbxt/K0Tzk2bsc76SSsRgrJSoh0fnzrB5M0Dqeqe7YHLxf1x7fU4S/FELV-cat.jpg",
  "image2": "https://replicate.delivery/pbxt/K0TzkJl3lco0aPVIg8iQYUB0ursK7ZWO0ECEVsxMMQlf5eKH/ironman.jpg",
  "prompt": "<i> in the suit of <i>",
  "negativePrompt": "",
  "numberOfImages": 1,
  "textGuidanceScale": 6,
  "numberOfInferenceSteps": 50
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json",
    # Add any other required headers for the Cognitive Actions API
}

# Prepare the request body for the hypothetical execution endpoint
request_body = {
    "action_id": action_id,
    "inputs": payload
}

print(f"--- Calling Cognitive Action: {action.name or action_id} ---")
print(f"Endpoint: {COGNITIVE_ACTIONS_EXECUTE_URL}")
print(f"Action ID: {action_id}")
print("Payload being sent:")
print(json.dumps(request_body, indent=2))
print("------------------------------------------------")

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json=request_body
    )
    response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully. Result:")
    print(json.dumps(result, indent=2))

except requests.exceptions.RequestException as e:
    print(f"Error executing action {action_id}: {e}")
    if e.response is not None:
        print(f"Response status: {e.response.status_code}")
        try:
            print(f"Response body: {e.response.json()}")
        except json.JSONDecodeError:
            print(f"Response body (non-JSON): {e.response.text}")
    print("------------------------------------------------")

Conclusion

Kosmos G offers developers a unique opportunity to harness the power of AI in image generation. With its ability to create contextually relevant images from multi-modal prompts, it simplifies the creative process while enhancing the quality of visual content. Whether you're looking to elevate marketing materials, enrich storytelling, or innovate in creative projects, integrating Kosmos G into your workflow can lead to impressive results. Start exploring the capabilities of Kosmos G today and unlock new avenues for expression and creativity!

← Back to All Articles