Imagen 3 Fast

Partner Model

Fast Inference

REST API

Model Information

Response Time:~10 sec

Status:Active

Version:

0.0.1

Updated:about 1 month ago

imagen-3-fast

Live Demo

Average runtime: ~10 seconds

Input

Configure model parameters

Negative Prompt

Text prompt for what to discourage in the generated images

Enter your negative prompt

prompt

Text prompt for image generation

Bullet train cutting through cherry blossom petals at dusk, creating a vortex of pink and white. The polished silver exterior reflects the setting sun in a continuous streak of fire. Station lights blend into a single continuous line as the train cleaves through space. The world outside the windows is nothing but beautiful motion blur.

aspect_ratio

An enumeration.

safety_filter_level

An enumeration.

Output

View generated results

Result

Preview, share or download your results with a single click.

Each execution costs $0.028 With $1 you can run this model about 35 times.

API Reference

View Full Documentation

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 requests
import time

API_KEY = "YOUR_API_KEY"  # Replace with your API key
HEADERS = {
    "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": "imagen-3-fast",
            "version": "0.0.1",
            "input": {
  "prompt": "your prompt here",
  "aspect_ratio": "1:1",
  "negative_prompt": "your negative prompt here",
  "safety_filter_level": "block_medium_and_above"
},
            "webhook_url": ""
        }
    )
    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 result
        elif 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 prediction
    prediction_id = create_prediction()
    print(f"Prediction created: {prediction_id}")
    
    # Get result
    result = 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: ~10 seconds
Rate limit: 60 requests/minute
Concurrent requests: 10 maximum
Use long-polling to check prediction status until completion

Overview

Imagen 3 Fast is a text-to-image generative model optimized by Google Deepmind for speed while maintaining high-quality image generation. It interprets textual descriptions to create visually compelling images with improved coherence, lighting, and texture. The Imagen 3 Fast supports multiple aspect ratios, fine-grained prompt control, and filtering options to customize generated outputs.

Technical Specifications

Architecture: Imagen 3 Fast is based on a diffusion model architecture optimized for rapid inference, balancing speed and quality.
Resolution: Generates images at a base resolution of 1024x1024 pixels, with support for upscaling.
Language Processing: Supports multiple languages for text input, allowing diverse linguistic accessibility.
Text Rendering: Can generate images containing text, though accuracy may vary in complex layouts.

Key Considerations

Prompt Clarity: More specific prompts lead to better image coherence. Vague prompts may yield unpredictable results.
Safety Filter Impact: Stricter filtering may prevent certain creative outputs. Adjust appropriately based on content needs.
Negative Prompts: Effective use of negative prompts can refine image composition by removing undesired elements.
Aspect Ratio Effects: Different aspect ratios impact how elements are arranged in an image. Choosing the right one improves visual balance.

Legal Information for Imagen 3 Fast

By using this Imagen 3 Fast, you agree to:

Google DeepMind Privacy
Google DeepMind TERMS

Tips & Tricks

Optimizing Prompts for Imagen 3:
- Use descriptive details such as colors, textures, and lighting conditions.
- Reference specific artistic styles (e.g., "watercolor painting," "cyberpunk style").
Negative Prompting for Refinement:
- If faces appear distorted, use negative_prompt like "blurred faces, distorted features."
- To remove specific backgrounds, include "cluttered background" in negative_prompt.
Aspect Ratio Selection:

1:1 – Best for balanced compositions or social media posts.
9:16 – Ideal for portrait-oriented images.
16:9 – Suited for landscape scenes or widescreen content.
3:4 and 4:3 – Useful for natural framing of objects and characters.

Fine-Tuning Safety Levels:
- If content is being blocked unexpectedly, lower the safety_filter_level cautiously.
- For stricter moderation, increase the filtering level.

Capabilities

Rapid Image Generation with Imagen 3: Produces high-quality images quickly based on text prompts.
Diverse Visual Styles: Supports multiple artistic styles and aesthetics.
Customizable Composition: Users can adjust aspect ratios, prompts, and negative prompts for tailored results.
Content Filtering: Allows control over content sensitivity with adjustable safety filters.

What can I use for?

Creative Content: Generate concept art, illustrations, and digital artwork.
Marketing Visuals: Create promotional materials, banners, and social media graphics.
Storytelling & Visualization: Develop visual assets for books, comics, and presentations.
Educational & Research Purposes: Generate instructional visuals or scientific illustrations.

Things to be aware of

Experiment with different aspect ratios to see how they affect composition.
Use detailed prompts with lighting descriptions for more realistic results.
Adjust negative_prompt to refine unwanted details in images.
Fine-tune the safety_filter_level to align with content needs.

Limitations

Complex Object Interactions: The Imagen 3 Fast may struggle with overlapping or highly detailed object relationships.
Text Generation Limitations: While improved, rendered text may not always be perfectly legible.
Potential Bias: Like other AI models, Imagen 3 Fast may reflect biases from its training data.

Output Format: PNG

Related AI Models

You can seamlessly integrate advanced AI capabilities into your applications without the hassle of managing complex infrastructure.

Imagen 4 Fast

Use this fast version of Imagen 4 when speed and cost are more important than quality

Stable Diffusion 3.5 Large

A text-to-image model that generates high-resolution images with fine details. It supports various artistic styles and produces diverse outputs from the same prompt, thanks to Query-Key Normalization.

Stable Diffusion 3.5 Medium

Stable Diffusion 3.5 Medium is 2.5 billion parameter image model with improved MMDiT-X architecture

Recraft V3

Recraft v3, code-named "red_panda," is a state-of-the-art text-to-image model designed to set a new standard in generating high-quality, realistic visuals for creative projects.