Fooocus
Model Information
Input
Configure model parameters
Output
View generated results
Result
Preview, share or download your results with a single click.
Prerequisites
- Create an API Key from the Eachlabs Console
- Install the required dependencies for your chosen language (e.g., requests for Python)
API Integration Steps
1. Create a Prediction
Send a POST request to create a new prediction. This will return a prediction ID that you'll use to check the result. The request should include your model inputs and API key.
import requestsimport timeAPI_KEY = "YOUR_API_KEY" # Replace with your API keyHEADERS = {"X-API-Key": API_KEY,"Content-Type": "application/json"}def create_prediction():response = requests.post("https://api.eachlabs.ai/v1/prediction/",headers=HEADERS,json={"model": "fooocus-api","version": "0.0.1","input": {"cn_stop2": null,"cn_stop3": null,"cn_stop4": null,"cn_weight2": null,"cn_weight3": null,"cn_weight4": null,"uov_input_image": "your uov input image here","loras_custom_urls": "your loras custom urls here","use_default_loras": "True","outpaint_selections": "your outpaint selections here","performance_selection": "Speed","cn_weight1": null,"sharpness": "2","image_seed": "-1","inpaint_input_image": "your inpaint input image here","refiner_switch": "0.5","cn_stop1": "0.5","uov_upscale_value": "0","image_number": "1","uov_method": "Disabled","outpaint_distance_left": "0","outpaint_distance_top": "0","guidance_scale": "7","inpaint_input_mask": "your inpaint input mask here","cn_type2": "ImagePrompt","cn_type4": "ImagePrompt","cn_type3": "ImagePrompt","prompt": "your prompt here","cn_img1": "your cn img1 here","cn_img2": "your cn img2 here","cn_img3": "your cn img3 here","cn_img4": "your cn img4 here","cn_type1": "ImagePrompt","style_selections": "Fooocus V2,Fooocus Enhance,Fooocus Sharp","negative_prompt": "your negative prompt here","inpaint_additional_prompt": "your inpaint additional prompt here","outpaint_distance_bottom": "0","outpaint_distance_right": "0","aspect_ratios_selection": "1152*896"}})prediction = response.json()if prediction["status"] != "success":raise Exception(f"Prediction failed: {prediction}")return prediction["predictionID"]
2. Get Prediction Result
Poll the prediction endpoint with the prediction ID until the result is ready. The API uses long-polling, so you'll need to repeatedly check until you receive a success status.
def get_prediction(prediction_id):while True:result = requests.get(f"https://api.eachlabs.ai/v1/prediction/{prediction_id}",headers=HEADERS).json()if result["status"] == "success":return resultelif result["status"] == "error":raise Exception(f"Prediction failed: {result}")time.sleep(1) # Wait before polling again
3. Complete Example
Here's a complete example that puts it all together, including error handling and result processing. This shows how to create a prediction and wait for the result in a production environment.
try:# Create predictionprediction_id = create_prediction()print(f"Prediction created: {prediction_id}")# Get resultresult = get_prediction(prediction_id)print(f"Output URL: {result['output']}")print(f"Processing time: {result['metrics']['predict_time']}s")except Exception as e:print(f"Error: {e}")
Additional Information
- The API uses a two-step process: create prediction and poll for results
- Response time: ~31 seconds
- Rate limit: 60 requests/minute
- Concurrent requests: 10 maximum
- Use long-polling to check prediction status until completion
Overview
This model is designed to generate high-quality, visually appealing images based on textual descriptions and additional user inputs. It supports advanced customization for style, aspect ratios, image refinement, and creative enhancements like inpainting, outpainting, and ControlNet integration. With an intuitive parameter structure, users can achieve precise results for various artistic and practical use cases.
Technical Specifications
- Architecture:
- Based on state-of-the-art deep learning frameworks for image synthesis.
- Utilizes an enhanced diffusion model for generating high-resolution visuals.
- Performance:
- Capable of producing images up to 1024x1024 pixels in resolution.
- Optimized for efficient processing with minimal latency.
- Customization:
- Supports fine-tuning for specific artistic styles or project requirements.
- Flexibility: Users can generate images in various artistic styles, refine outputs with additional prompts, and leverage control parameters for fine-tuning.
- Input Variety: The model supports a wide range of input types, including text prompts, images, masks, and numerical parameters.
Core Functionalities
- Image Generation:
- Generate unique, high-resolution images from detailed prompts.
- Adjust style, sharpness, and guidance settings for custom outputs.
- Inpainting and Outpainting:
- Modify specific areas of an image using masks and additional prompts.
- Extend image boundaries seamlessly with configurable distances for each side.
- ControlNet Integration:
- Allows precise control of image generation by integrating external guidance inputs, such as reference images or structural data.
- Refinement Options:
- Enable or disable refinement processes to balance output quality with performance speed.
- Aspect Ratio Customization:
- Supports predefined and custom aspect ratios, ensuring compatibility with various use cases.
- Style and Performance Modes:
- Select predefined styles for artistic effects.
- Optimize performance based on speed or quality requirements.
Prompt Clarity: Craft descriptive and specific prompts to guide the model effectively. Avoid overly ambiguous or contradictory instructions.
Negative Prompting: Use the negative_prompt to exclude unwanted elements (e.g., "no text, no watermarks").
Style and Aspect Ratios: Select styles and aspect ratios carefully to align with the intended visual outcome.
Image Refinement: Use the refiner_switch to adjust sharpness or enhance details in generated images.
ControlNet Features: Include ControlNet images (cn_img1, cn_img2, etc.) to incorporate structured guidance, like specific poses or layouts.
Consistency: Use fixed seeds (image_seed) for reproducible outputs.
Customization: Combine loras_custom_urls with use_default_loras to apply pre-trained or custom LoRAs for style and content adjustments.
Key Considerations
- Parameter Synergy: Combine guidance_scale, sharpness, and refiner_switch for balanced quality and prompt adherence.
- Image Inputs: Use high-resolution input images for inpainting and outpainting tasks to maintain quality.
- Prompt Alignment: Ensure prompts match your desired style and subject to avoid conflicting results.
- ControlNet Layers: Use ControlNet inputs strategically for structured outputs, limiting layers to essential elements to prevent overloading.
Tips & Tricks
- Experiment with Styles: Test different style_selections to find the best fit for your project.
- Negative Prompt Precision: List specific elements to exclude for cleaner results.
- Custom LoRAs: Combine default and custom LoRAs for unique outputs.
- Outpainting Adjustments: Set outpaint_distance_* values to extend images seamlessly.
- ControlNet Fine-Tuning: Adjust cn_weight* and cn_stop* for subtle or pronounced effects.
Capabilities
- Artistic Image Generation:
- Creates visually stunning artwork in various styles, from realistic to abstract.
- Customizable Outputs:
- Tailors images to specific themes, moods, or artistic techniques.
- High-Resolution Visuals:
- Produces professional-quality images suitable for printing or digital use.
- Creative Experimentation:
- Encourages exploration of unique ideas and concepts.
What can I use for?
- Creative Projects:
- Develop concept art, illustrations, or digital paintings.
- Design Work:
- Enhance branding, marketing materials, or website visuals.
- Educational Content:
- Generate visual aids for presentations, tutorials, or storytelling.
- Personal Use:
- Create custom wallpapers, gifts, or personal art collections.
Things to be aware of
- Theme-Based Prompts:
- Experiment with themes like "futuristic cityscape," "fantasy forest," or "minimalist design."
- Artistic Styles:
- Explore styles such as watercolor, charcoal sketch, or surrealism.
- Blend Ideas:
- Combine contrasting concepts for unique results (e.g., "ancient ruins in a cyberpunk world").
- Iterative Refinement:
- Generate multiple versions of an image and select or combine elements you like.
Limitations
- Complex Prompts: Overly detailed or conflicting prompts may reduce output quality.
- Resolution Constraints: Extremely high resolutions may increase processing time.
- ControlNet Dependencies: Overusing ControlNet layers can complicate outputs and slow processing.
- Mask Accuracy: Poorly defined masks in inpainting tasks may lead to unintended edits.
Output Format:PNG
Related AI Models
Flux Dev
flux-dev
Sana by Nvidia
sana
Ideogram V2
ideogram-v2
