How To Use Flux Kontext In Flux Labs AI
Black Forest Lab's Flux Kontext image model is now supported in Flux Labs AI. Here's how you can use it and some cool use-cases you can try.
Black Forest Labs’ latest suite of image models, Flux Kontext, is now supported in Flux Labs AI.
By default, Flux Kontext will be powering the Image Manipulator feature. Simply upload an image and reimagine it by describing the changes you want. You can change a portion of it or change the whole thing. You can use it to transfer an image style or restyle the image itself. There are so many things you can do with this model. It’s versatile, powerful, and super fun to use.
In this guide, I’ll be showing you how you can use Flux Kontext in Flux Labs AI. I’ll show you some awesome examples and cool tricks you can do with this new model. It’s so powerful, you won’t even have to use Photoshop or any other image editing tool anymore.
But before we dive into the examples, let’s review what’s new in Flux Kontext.
What is Flux Kontext?
Flux Kontext is a suite of generative flow matching models that allows you to generate and edit images. Unlike existing text-to-image models, the Flux Kontext family performs in-context image generation, allowing you to prompt with both text and images, and seamlessly extract and modify visual concepts to produce new, coherent results.
There are three models, two of which are available right now, while the other is in private beta:
Flux Kontext Pro: First model to build upon previous edits while maintaining characters, identities, styles, and features consistent across scenes
Flux Kontext Max: Maximum performance at high speed — improved prompt adherence, typography & premium consistency
Flux Kontext Dev (available in Private Beta): A lightweight 12B diffusion transformer suitable for customization and compatible with previous Flux 1.0 Dev inference code
Here are the core capabilities of the Flux Kontext model suite:
Character consistency: Preserve elements across scenes
Local editing: Target specific parts without affecting the rest
Style reference: Generate new scenes in existing styles
Interactive speed: Iterate with minimal latency
It also allows you to iteratively add more instructions and build on previous edits, refining your creation step-by-step with minimal latency, while preserving image quality and character consistency.
You can learn more about the performance details of Flux Kontext and some example images in this article.
How To Use Flux Kontext in Flux Labs AI
Head over to Flux Labs AI and create an account for free. Under the AI Tools dropdown menu, select the “Image Manipulator” and you will be redirected to the image editing page.
In the image editing page, you need to upload the input image. Make sure that it does not exceed 3 MB in size and is only one of the following formats: JPG, PNG, or WEBP.
Before showing you an example, take note of the available settings you can change. Aside from the text prompt, you can also change the output image aspect ratio and the model to be used.
Google Gemini: This model uses Gemini Pro’s image-to-image capability.
Flux Kontext Pro: This is the latest model from Black Forest Labs that can manipulate an image through text descriptions.
Flux Kontext Max: Same as the Pro version, but more powerful and more expensive.
Let’s try to turn this image of an asian woman into a snow-filled street using this prompt:
Prompt: It’s now snowing, everything is covered in snow
For this example, I set the aspect ratio to 16:9 and use the Flux Kontext Max model. Finally, click on the submit button and wait for the final image to be generated. Here’s the final result:
First of all, the image generation is fast. This example only took 3 seconds to complete. Second, the quality is superb. Look at how accurate and realistic those snow details are. Even the wheel tracks on the road are incredibly well rendered.
You can continue the workflow by describing further changes you want to make to the AI. Let’s say we want to change the attire to a leather jacket. Let’s set this description:
Prompt: Turn ger sweatshirt into a black leather jacket
Awesome. The jacket was placed seamlessly into the subject, and all other elements of the photo were retained. I am loving the results so far.
Here’s a fun little trick you can do on Flux Labs AI. If you are planning to post the image on social media, you can transform the image into a 9:16 format and turn it into a video. Here’s how you can do it:
First, set the aspect ratio to 9:16 and describe any changes you want. Say we want her to be wearing a red shirt, here’s the prompt:
Prompt: she’s wearing a red shirt under the leather jacket
Great. Now download the image to your local disk by clicking on the little download button on the upper right side.
Next, go to the UGC video tool page and do the following:
Check the “No need for product photo” to disable the product integration
Upload the previously downloaded image to the AI influencer section
Disable the voice narration
Set the duration to 10 seconds
Set the video quality to Ultra HD
Prompt: A girl showing excitement being in the snow, she’s smiling and moving to show the beautiful place to the camera
Once all the settings and description are set, click on the Generate button and wait for the final video to be rendered.
Here’s the side-by-side result of the image and the video:
This is really cool. The workflow showed how versatile the image editor tool is. Aside from changing the environment, attire, and aspect ratio, there are a lot more possible use cases with Flux Kontext. Let’s discuss some of them in the next section.
Fun Use Cases
Here are some of the most fun use cases I discovered where Flux Kontext shines.
First, it works well with restoring old photos and colorizing grayscale photos. Just upload the image and tell the AI to add appropriate colors to it.
Another cool trick you can do with the model is to change the aspect ratio. No, we’re not talking about cropping the input image. The AI is smart enough to seamlessly outpaint the empty spaces of the image frame.
To do that, just change the aspect ratio in Flux Labs AI and describe the elements to be added in the empty spaces, or simply describe it as “make the image taller or wider”.
Is this the end of watermarked photos? Flux Kontext can also remove watermarks or any text on a photo with just a simple instruction. The example below shows not just a watermark but a big layer of text overlays.
The result is a really well-edited version with no trace of the overlay. I mean, look at the new image, there are no weird artifacts that remained, no matter how much I zoomed in. This really blows my mind.
It also works great with texturizing or previewing the final render of a 3D object. Check out this example:
These are just some of the most fun and useful ways Flux Kontext can be used. In the coming weeks, Flux Labs will have support for multiple image inputs and custom apps, such as cartoonify, change haircut, professional headshots, and more.
Why Should You Care?
The image editing game just changed completely. While Google Imagen and GPT-Image-1 were decent for basic text-based image editing, Flux Kontext absolutely destroys them in every way possible.
I’ve been testing both models extensively over the past few days, and honestly, this is the most impressive image model I’ve encountered in years. Black Forest Labs has this incredible track record of releasing models that consistently exceed expectations, and Flux Kontext continues that streak.
Now, what makes this release extra cool is the Flux Kontext Dev that it’s open-weights. This means you can actually download the entire model to your computer and run it using your own hardware. No internet connection required, no subscription fees, and complete privacy for your projects. It’s cheap and secure, plus it opens the door for the community to create fine-tuned versions for specific use cases.
As regular users, we shouldn’t be locked into the ecosystems of closed-source companies like Google and OpenAI. As someone who writes about AI and uses these tools daily, I strongly encourage people to consider Flux as their primary option. I don’t even think it’s an alternative. In many cases, it actually outperforms the so-called industry leaders.
Final Thoughts
Flux Labs AI just became significantly more powerful with Flux Kontext support. I’ve spent time with Gemini Flash’s image generation and OpenAI’s GPT-Image-1, but Flux Kontext completely overpowers both of them. The interface couldn’t be simpler either. Upload your photo, describe what you want changed, pick your aspect ratio, and hit generate. That’s it.
What gets me really excited is what’s coming next.
The developers mentioned they’re rolling out support for multiple image uploads and dedicated Flux Kontext apps in the near future. I can already imagine the possibilities this will unlock for the Image Editing tool.
You need to try this feature yourself on Flux Labs AI. Seriously, go play with it for an hour and see what you can create. Then let me know what you think about it in the comments.
Hi there! Thanks for making it to the end of this post! My name is Jim, and I’m an AI enthusiast passionate about exploring the latest news, guides, and insights in the world of generative AI. If you’ve enjoyed this content and would like to support my work, consider becoming a paid subscriber. Your support means a lot!