Recraft V3

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.

Fast Inference

REST API

Try in Console API Docs Examples

Model Information

Response Time~11 sec

StatusActive

Version

0.0.1

Updated20 days ago

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": "recraft-v3",
            "version": "0.0.1",
            "input": {
  "size": "1024x1024",
  "style": "any",
  "prompt": "your prompt here"
}
        }
    )
    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: ~11 seconds
Rate limit: 60 requests/minute
Concurrent requests: 10 maximum
Use long-polling to check prediction status until completion

Overview

Recraft v3 is designed to generate high-quality, customizable images tailored to your specific parameters. By adjusting various inputs, you can create images that align with your vision—whether you're looking for realistic visuals, artistic flair, or a unique structure. Its intuitive parameter system makes it easy to use and adaptable for a wide range of creative projects, providing maximum flexibility and customization.

Technical Specifications

Key Features for Recraft v3:

Dynamic Prompting: Generates images based on text descriptions for unmatched creativity.
Resolution Flexibility: Offers a wide range of resolutions for various use cases.
Stylistic Versatility: Includes predefined styles to accommodate diverse artistic preferences.

Input Parameters:

prompt
- Represents the text input to guide image generation.
- Example: "A serene forest during sunrise" or "A futuristic cityscape with neon lights."
size
- Defines the resolution of the output image.
- Supported values:
  - Landscape: 1024x2048, 2048x1024
  - Portrait: 1024x1365, 1024x1536, 1024x1820
  - Square: 1024x1024, 1280x1024
style
- Determines the visual aesthetic of the output.
- Categories include:
  - Realistic: natural_light, studio_portrait
  - Artistic: pixel_art, engraving_color, handmade_3d
  - Mixed styles: b_and_w, hard_flash, motion_blur

Key Considerations

The prompt must align with the desired style and resolution for optimal results for Recraft v3. Avoid contradictions, such as requesting a realistic image with an overly abstract prompt.

The size significantly impacts processing time. Larger sizes may increase the time required but produce highly detailed results.

Style combinations are predefined. Use them to match the desired visual theme, but be mindful of potential mismatches with the prompt.

Legal Information for Recraft V3

By using this model, you agree to:

Recraft Privacy
Recraft Terms

Tips & Tricks

Prompt:
- Include clear, concise descriptions with specific elements.
  - Good: "A vintage car parked under autumn trees with vibrant leaves."
  - Avoid: "A car under trees" (too vague).
- Add modifiers for enhanced detail, such as "sunlit," "foggy," or "vibrant."
Size:
- High Detail Outputs: Use resolutions like 2048x1024 or 1024x2048 for maximum clarity in large scenes.
- Quick Previews: Select smaller resolutions such as 1024x1024 to reduce generation time.
- Platform-Specific: For social media, use resolutions like 1024x1280 or 1280x1024.
Style:
- Realism: Use natural_light or studio_portrait for lifelike outputs.
- Artistic Touches: Experiment with pixel_art, hand_drawn, or engraving_color for creative variations.
- Dynamic Scenes: Try motion_blur for action-packed or fast-moving visuals.
- Avoid overusing niche styles like infantile_sketch unless specifically needed for a whimsical effect.

Capabilities

You can generate photorealistic or artistically rendered images with Recraft v3.

Offer a wide variety of stylistic options for creative exploration.

Adjust outputs based on specific resolution needs.

What can I use for?

Creative Projects for Recraft v3 : Design unique illustrations for personal or professional use.

Content Creation: Produce visually stunning images for blogs, presentations, or media.

Prototyping: Quickly visualize ideas or concepts.

Things to be aware of

Combine realistic styles with high resolutions to create lifelike portraits or landscapes.

Explore artistic styles like handmade_3d or engraving_color for a distinctive aesthetic.

Use motion_blur for dynamic, high-energy scenes that simulate movement.

Experiment with different size configurations to find the ideal balance between detail and file size.

Limitations

Style Overlap: Combining multiple stylistic categories may produce inconsistent results.

Complex Prompts: Overly detailed descriptions might not always translate accurately into the generated image.

Resolution Limitations: Very large sizes may strain resources or impact generation speed.

Output Format: WEBP

Related AI Models

Flux 1.1 Pro

flux-1-1-pro

Text to Image

Stable Diffusion 3.5 Medium

stable-diffusion-3-5-medium

Text to Image

Fooocus

fooocus-api

Text to Image

Photon Flash

photon-flash

Text to Image