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": "multi-image-kontext-pro","version": "0.0.1","input": {"seed": 0,"prompt": "your prompt here","aspect_ratio": "match_input_image","input_image_1": "your input image 1 here","input_image_2": "your input image 2 here","output_format": "png","safety_tolerance": 2},"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 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: ~15 seconds
- Rate limit: 60 requests/minute
- Concurrent requests: 10 maximum
- Use long-polling to check prediction status until completion
Overview
Flux Kontext Pro Multi Image is an advanced multi-image conditioning model that combines two reference images with a natural language prompt to generate high-quality, context-aware visuals. It is designed to extract nuanced visual and stylistic cues from each image and harmonize them with the given textual instruction to produce coherent and detailed output.
Technical Specifications
Architecture: Flux Kontext-based dual-conditioning generation
Conditioning Sources: Dual image encoding + prompt fusion
Output: Single image conditioned on two reference inputs and text
Fine-tuned for consistency, facial structure retention, and context blending
High compatibility with variable aspect ratios and safety parameters
Key Considerations
Overly abstract prompts without clear visual guidance from images may lead to unpredictable outputs.
Avoid using heavily stylized or inconsistent images together, as Flux Kontext Pro Multi Image might fail to establish visual harmony.
The safety_tolerance slider can restrict certain visual generations; lowering it too much may remove essential features, while raising it too high may lead to risky or distorted results.
The seed value, if fixed, can help with reproducibility, but will not guarantee identical results in all scenarios due to stochastic processes.
Tips & Tricks
Prompt
-
Use direct, descriptive prompts that align with the context of both input images.
Example:
✅ “A cinematic portrait of a woman in a futuristic city, inspired by both images”
❌ “Dreamy vibes with some magic” - Avoid flagged or sensitive terms. Words referring to violence, nudity, explicit content, or politically charged subjects may be blocked or filtered.
Input Images
- Recommended resolution: Minimum 512x512px for each image.
- Images should contain the subject in similar lighting or perspective for optimal blending.
- Avoid using extremely cluttered or low-resolution images; they reduce feature extraction accuracy.
Aspect Ratio
- match_input_image: Retains the layout of the first input image
- 1:1: Balanced framing
- 16:9 or 21:9: Best for cinematic compositions
- 3:4 or 4:5: Useful for portrait-style outputs
- Aspect ratio mismatches between input and output can lead to subject distortion
Seed
- Use a fixed integer (e.g., 1234) for reproducible results.
- Changing the seed gives a new variation on the same inputs and prompt.
Output Format
- jpg: Smaller size, slightly compressed quality
- png: Larger file, sharper quality — ideal for design or print
Capabilities
Dual image fusion with accurate context extraction
Generates images that maintain visual consistency across source images
Supports detailed prompts and flexible aspect ratios
Enables image style transfer, character preservation, and layout guidance
Supports reproducible outputs via seed control
What can I use for?
Creating visual references or storyboards using dual image and text cues
Generating character or concept art using multiple references
Combining visual identities (e.g., outfit + face) to generate coherent portraits
Enhancing creative workflows that need high control over stylistic elements
Things to be aware of
Combine a celebrity face with a fashion photo to create styled character designs
Use landscape photography + painting as inputs to generate fantasy scenes
Experiment with 3:2 or 4:5 ratios for editorial-style compositions
Adjust safety tolerance and compare the visual clarity of different outputs
Limitations
Outputs may struggle with complex prompt logic when reference images conflict
Some extreme aspect ratios can introduce composition issues
Flux Kontext Pro Multi Image does not support animation or video generation
Safety filters may limit creative freedom in certain themes
Cannot guarantee perfect pose transfer or spatial consistency between inputs
Output Format: JPG,PNG
Related AI Models
You can seamlessly integrate advanced AI capabilities into your applications without the hassle of managing complex infrastructure.
Eachlabs Background Remover v1
Eachlabs Background Remover v1 a reliable model that accurately removes backgrounds from images, making it easy to isolate subjects for product showcases, design work, or clean visual presentations.
GPT-1 Image Edit
OpenAI Image Edit lets you modify images by removing, adding, or changing parts. It uses AI to fill in the selected area naturally.
Custom Image Generation v2
By combining an input image with preset styles, Custom Image Generation v2 generates refined images that retain the original layout with stylistic updates.
Flux Canny Pro
Flux Canny Pro Model delivers accurate edge detection for precise outlines and detailed image edits