Flux.1 Kontext Max

Fast Inference

REST API

Model Information

Response Time:~0 sec

Status:Active

Version:

0.0.1

Updated:6 days ago

flux-kontext-max

Live Demo

Average runtime: -

Input

Configure model parameters

Prompt

Text description of what you want to generate, or the instruction on how to edit the given image.

Change the cat’s and dog’s color to white

Input Image

Image to use as reference. Must be jpeg, png, gif, or webp.

File upload is currently disabled

Output

View generated results

Result

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

Each execution costs $0.08 With $1 you can run this model about 12 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": "flux-kontext-max",
            "version": "0.0.1",
            "input": {
  "safety_tolerance": 6,
  "seed": 0,
  "prompt": "your prompt here",
  "input_image": "your input image here",
  "aspect_ratio": "match_input_image"
},
            "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: -
Rate limit: 60 requests/minute
Concurrent requests: 10 maximum
Use long-polling to check prediction status until completion

Overview

Flux.1 Kontext Max is an advanced single-image transformation model designed to reinterpret visual concepts by merging prompt-driven direction with detailed image understanding. It excels in enhancing, reimagining, or evolving existing imagery based on semantic intent, offering a flexible and prompt-guided creative process. Flux.1 Kontext Max balances prompt influence with image fidelity, providing refined outputs aligned with artistic or conceptual goals.

Technical Specifications

Native support for multi-aspect decoding enables dynamic reshaping of the output canvas based on selected ratios.

Flux.1 Kontext Max has been optimized for semantic stability across multiple iterations, maintaining object identity and spatial alignment even under extensive prompt changes.

Key Considerations

Prompt and input image should conceptually align. If not, results may appear incoherent or visually mismatched.

Very abstract or contradictory prompts can reduce generation quality.

Extreme aspect ratios (like 21:9 or 1:2) may introduce artifacts or excessive stretching.

Safety filters are not designed for fine-tuning outputs. For best visual quality, pair with a suitable prompt and a moderate safety setting.

Outputs are not intended for photorealistic document rendering or strict scientific visualization.

Tips & Tricks

Prompt
Use 7–15 word prompts. Example: "A serene lakeside landscape at sunset with soft fog and reflections."
Avoid vague terms like “cool” or “nice”; be specific with elements, style, and mood.

Input Image
Use images with a clear subject, clean composition, and moderate contrast. Avoid cluttered or noisy visuals.

Aspect Ratio

match_input_image keeps the original aspect and is safe for most cases.
1:1, 4:5, and 3:4 work well for portraits.
16:9, 21:9, or 3:2 are suitable for landscape or cinematic results.
2:3 and 1:2 can create vertically framed artworks but may crop top/bottom.

Seed

Fixed seed (e.g., 42) gives consistent results for the same input.
Leave blank for different results each time.
Changing the seed slightly (e.g., 40, 41, 42) offers subtle variations.

Safety Tolerance

1: Strictest filtering (highly safe but may alter artistic intent).
3: Balanced creative output with low risk.
6: Highest freedom, use when more experimental visuals are desired.

Capabilities

Transform real images based on creative or stylistic prompts.

Modify scenery, lighting, weather, and atmosphere in images.

Extend, recompose, or conceptually reframe existing visuals.

Preserve structure and perspective while updating content semantically.

Maintain facial and object identity across style or context changes.

What can I use for?

Creating thematic illustrations from reference photos.

Stylizing existing product shots or concept images.

Preparing mood boards or visual drafts for design direction.

Generating editorial visuals guided by text descriptions.

Exploring alternate visual interpretations of a single image.

Things to be aware of

Input a neutral landscape photo and prompt it with different times of day: "sunset", "night with stars", "early morning fog".

Reimagine a portrait with style prompts: "cyberpunk look with neon lights", "painted in Van Gogh style", "1920s noir photo".

Experiment with aspect ratios: Try 16:9 for cinematic scenes, 1:1 for posters, 4:5 for vertical editorial mockups.

Adjust the seed while keeping all else the same to explore alternate moods or visual interpretations.

Limitations

Limited control over small details or specific object positioning.

Very long or ambiguous prompts may reduce visual clarity.

Output is influenced by both the prompt and the image; unexpected results may occur if they are misaligned.

Extreme aspect ratios may result in less refined compositions.

Not suitable for highly technical illustration needs where accuracy is critical.

Output Format: PNG

Related AI Models

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

Eachlabs | AI Workflows for app builders

Flux Multi Image Kontext

An experimental FLUX Kontext model that can combine two input images

Product Shoot

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.

Eachlabs Face Swap

Swap faces between images effortlessly! Integrate your app with the Faceswap model by Eachlabs for smooth and seamless transformations.