Aderek Flux FP8 v2
v2















By creating this version, I've corrected the errors found in the previous ones. The model is a combination of various LoRAs (mine, x-lab ai, and others) merged with a clean FLUX.1 D checkpoint. It was subsequently converted to fp8 format to allow more people to download and run the model locally (and also for faster cloud-based operation). This model was designed with the goal of creating photorealistic science fiction and fantasy images. However, thanks to a delicate combination of multiple LoRAs, the model also excels at replicating manga, cartoons, various landscapes, and even the Midjourney style. The LoRAs used include: adimaBokeh-Flux-V0.1 (source unknown), flux-warrior (trained on my best SDXL and Stable Diffusion images), AngelDreamFLux (also mine, trained on SDXL and DALL-E images), and LoRAs from X-LABS - realism, anime, art, mjv6, scenery (https://huggingface.co/XLabs-AI/flux-lora-collection/tree/main). I encourage you to use this checkpoint to generate your own amazing images :)
How ?
I use only comfyui on my owm PC.
Pleas visit my sites:
https://www.deviantart.com/aderek
https://www.instagram.com/tomczuk.pawel/
https://stock.adobe.com/pl/contributor/211456689/aderek514?load_type=author&prev_url=detail
Discussion
Gallery92


























