SeedVR2 Upscaler for Video and Images

SeedVR2 Upscaler for Video and Images

SeedVR2 is an upscaler model by ByteDance-Seed for Video Restoration and is licensed under Apache-2.0 license. This implies that can be used in commerical capacity if you want. The one thing to note is that this model is GPU VRAM hungry model and you will struggle to...
How to Stylize Your Images with Flux Kontext in ComfyUI

How to Stylize Your Images with Flux Kontext in ComfyUI

In this post we will explore how we can use Flux Kontext to stylize images inside ComfyUI. Whether you’re working with your own photography or AI-generated images, Flux Kontext lets you apply different artistic styles effortlessly. Think of it like giving your images...
Flux Kontext can create LoRA Dataset

Flux Kontext can create LoRA Dataset

Flux release today to the public their Flux Kontext model which is designed to replace many different tools we’ve used like Inpainting, outpainting, controlnet, LoRA and its newest yet most powerful feature in context image editing with simple text prompts. No...
LoRA Training with AI-Toolkit on Vast.Ai under 50 cents

LoRA Training with AI-Toolkit on Vast.Ai under 50 cents

After exploring may of the possibilities on the RunPod platform, I started to explore other providers. Vast.Ai offers RTX4090 for under $0.49 per hour which means that its 20 cents cheaper than using it on RunPod. So as my first experiment I decided to see if I can...
F-Lite AI new model by Freepik and Fal

F-Lite AI new model by Freepik and Fal

A new model in the AI world has been release just two days ago. It is called F-Lite AI which has been released by Freepik and Fal, there is actually two of them F-lite and F-Lite-Texture; they have produced these as open source models but I assume that Lite means...

4x Upscale with AuraSR

There are plenty of Upscalers available in the open source space and in this post we explore AuraSR which is released by Fal-AI which has been incorporated into ComfyUI using the custom node AuraSR-ComfyUI. This upscaler is a GAN-based Super-Resolution for real-world...