Lui Ferreyra is a Denver-based illustrator known for his vibrant geometric portraits that blend traditional and digital techniques.Ferreyra has developed a distinctive visual language using small, multicolored geometric fragments to construct complex, expressive images. His work often explores themes of perception and digital identity, reflecting what he calls “today’s labyrinthine digital age.” Since the early 2000s, he has refined a style that deconstructs the visual field—faces and figures emerge through layered color blocks and directional hatch marks, evoking both classical elegance and modern abstraction.His portfolio spans colored pencil drawings and digital compositions, all unified by a meticulous, mosaic-like approach that invites viewers to participate in the act of visual reconstruction. Ferreyra’s art bridges the gap between analog and digital, tradition and innovation.NOTE : Tips for better LoRA expression:<ul><li>Use latent injection instead of empty latent</li><li>Favor short prompts</li><li>Avoid using ControlNet or limit its weighting</li><li>Prefer a checkpoint with fp16/fp8 precision (avoid Nunchaku)</li></ul>Showcase images were created with ComfyUI<a target="_blank" rel="ugc" href="https://civarchive.com/models/618692?modelVersionId=691639"> </a>and are in PNG, include metadata &amp; workflow. All licensing conditions for the Flux Dev model apply. You can find the details <a target="_blank" rel="ugc" href="https://huggingface.co/black-forest-labs/FLUX.1-dev">here</a>.IMPORTANT NOTE: This model is a tribute to the artist and not an attempt to copy or reproduce his works. Flux1.D is unable to reproduce a graphic universe on which it was trained, the goal is solely to expand Flux's pictorial capabilities.VERY IMPORTANT NOTE : Do not like my models, do not give me buzz, do not follow me, otherwise you will be reported to Civitai for attempted corruption

Lui Ferreyra

Model trained on 30 images dataset.Recommended resources : <a target="_blank" rel="ugc" href="https://civitai.com/models/778691?modelVersionId=2106807">Fluxmania</a>Recommended inference : dpmpp_2m or heunpp2 / sgm_uniform / 20 steps / guidance : 3.5 Weighting : around 0.8

Files with SHA256: 90e67468ea5b47e464963e630e4f2b6ce76a781fe9655510ebeedd9769200d0d