Nice to meet you all! I'm Alessandro Delfino, italian graphic design professor and AI enthusiast. As a traditional illustrator, I really felt like something amazing was happening as soon as I started to generate pictures with AI - it was like Christmas times eleven for me... but the process was pretty rough. _______ 😬 "What was my workflow with the good old Stable Diffusion 1.4?" Simple, once I found an interesting composition, my workflow was something like this: 1- Use that image as a starting point for image-to-image generations 2- Adjust the prompt as needed and generate 4-6 variants 3- Select the preferred image among them and start over from step 1 It was a continuous process of reworking images, correcting and modulating, which allowed me to overcome the great limitations of that old model and generate something close to what I wanted, one step at a time... The problems always started when I liked two images or I was curious about what a specific image would become if I made some changes to its prompt... At that point I had to start branching the process, remembering the origin of each generation and copying all the prompts somewhere, trying to keep track of the variations... a process that led to great results, at the mere cost of my mental health - especially when you think that the only way to not lose that data was to keep my PC running overnight. _______ 🤩 "If it doesn't exist...build it!" I was lucky enough to have friends like Alberto Castella, which is an expert in complex systems and an amazing programmer, & Samuel Suni a young and amazing UI designer that had previously worked on EasyDiffusion and was now hooked on that approach to AI images - after sharing with them that problem and the solutions I had in mind, we all started to work full-time on the app which is now called Arthemy. Arthemy is an AI powered image generator which I feel confident enough to define "professional" thanks to its visualization of those bifurcations with a reliable visual map that would allow me to focus on the results, without being worried about what I leave behind. I didn't have to trace maps anymore, now I had GPS. Which is great for businesses and personal projects, because you can store your ideas and evolve them as many times as you want, it's a creative tool that helps to think by seeing. _______ 😎 "Arthemy is not a tool, it's a process" Nobody can really predict what AI will be tomorrow, and during revolutions of this magnitude every software seems to be unbeatable... for about two weeks. For this reason, we have developed an autonomous interface which will grants us the possibility to change the generative engine when something revolutionary will come. _______ 🧙‍♂️ "Who is arthemy for?" It's pretty much for all people who wants to... - Have a history of their generations (both for the valorization of the final product and for a greater freedom of experimentation) - Work on multiple images at the same time, without getting lost. - Change their generation model for total customization of the outputs - Save their process and continue the next day, without losing anything - Store projects and results by client or commission, making them accessible at all times - Improve your marketing campaigns thanks to some very neat A/B testing with slightly different pictures. _______ But we're not just that, we want to build a whole world of trained models, trained on the style of specific artists or photographers - with their consent - making those models premium and sharing the profit with the artists themselves! We have a lot of ideas and we can't wait to discuss with any of you on our social networks: https://linktr.ee/arthemyai Join now and try arthemy at https://app.arthemy.ai

arthemy

AI image generator for professional creatives

AI image generator for professional creatives