Magnific AI

⚡ How create hyper realistic food photography with Generative AI in 4k resolution ready to print (4,736 x 3,520 px) Step-by-step tutorial 🧵👇 First things first: all the generations in this thread follow these general parameters in Magnific Mystic v2 (tweaked for food photography). Make sure to set them this way: - 4k resolution - Creative: 85% Check out the screenshot in the gallery! Prompts for Magnific Mystic v2: - Food photography, creamy Panna Cotta garnished with fresh berries, soft lighting, close-up shot, photorealistic, 8k render, elegant dessert plate on white background - Food photography, juicy Cheeseburger with crispy bacon and melted cheddar, dramatic lighting, shallow depth of field, photorealistic, 8k render, urban diner setting - Food photography, Japanese ramen bowl with pork belly, soft-boiled egg, and noodles, studio lighting, photorealistic, 8k render, minimalist presentation - Product photo of a grilled trout in a dish with bacon, smoldering coal embers, dark background, studio light, photorealistic, hyperrealistic, 8K - Food photography, Crème Brûlée with caramelized sugar crust, warm lighting, close-up, photorealistic, 8k detail, fine dining table setting - Food photography, seared scallops with truffle foam and microgreens, studio lighting, photorealistic, 8k resolution, minimalist white background - Food photography, Chocolate Lava Cake with molten center oozing out, low-key lighting, close-up shot, photorealistic, 8k detail, dark rustic background - Food photography, passion fruit soufflé dusted with icing sugar and mint garnish, studio lighting, photorealistic, 8k quality, fine dining dessert presentation - Food photography, crispy skin salmon with beetroot purée and dill sauce, studio lighting, photorealistic, 8k render, elegant plating on white dish - Food photography, homemade Apple Pie with lattice crust, warm lighting, overhead shot, photorealistic, 8k resolution, country kitchen background - Food photography, Japanese sushi assortment with nigiri and maki rolls, studio lighting, photorealistic, 8k resolution, minimalist white background - Food photography, asparagus velouté with poached quail egg and truffle oil, studio lighting, photorealistic, 8k clarity, gourmet soup presentation - Food photography, tuna tartare with wasabi aioli and sesame tuile, studio lighting, photorealistic, 8k detail, modern plating on black plate - Food photography, roasted quail with grape jus and wild rice, studio lighting, photorealistic, 8k quality, elegant dish on rustic background - Food photography, slow-cooked pork belly with apple purée and crackling, studio lighting, photorealistic, 8k clarity, fine dining plating You can check out the rest of the prompts here: https://lnkd.in/gpq_ED2p Enjoy! 📌 Tool and param details in screenshots. ➕ Follow for more content! #ai #generativeAI #photography #magnific #genai

As they say in Korea, 감사합니다 🤞 Our wonderful Paula Vivas, Head of Marketing in the U.S., along with our partners Javi López and Emilio Nicolás Riquelme from Magnific AI, were invited to the GenAI Innovation Conference in Seoul. They took the stage to discuss generative AI, the collaboration between our two brands, Freepik Mystic, and the future of the image and design industry. "Our model has a 90% precision in English, Spanish, French, and Italian," shared Javi López, "[...] and we're working to improve support in Korean and Japanese." It was an incredible experience, and we can't wait to see you all again. 📰 You can read more about the event here: https://lnkd.in/dvPVCisT

Grab pen and paper (if you're still using one of those), and take some notes. These are things you can do in Magnific that you might not know. Let's go! ⭐ Reuse parameters. Reusing all the configurations of a previous job and/or the original or final image. ⭐ Add to the queue. The infra will handle them in parallel, so you can generate things at the speed of light. ⭐ Copy any previous prompt used just with a click. ⭐ Zooming/moving any result with Z + mouse. And grid and faves, but you probably know those.

Last week, we launched Mystic V2 on Magnific AI, and we’re already blown away by the creations our X closest friends had came up with 🔥 These are some of the jaw-dropping designs we’ve spotted so far. And don’t stop now—keep tagging us in your creations so we can keep the spotlight on your amazing work. 🚀✨ Swipe to the left, and comment you favorite, because we can't decide 👇💬

🔥 Mystic v2 is out! 🔥 Magnific is no longer just an upscaler, it’s now one of the top AI image generators out there, creating images up to 4K quality. This isn’t just an update; it’s a fresh start for us.

https://lnkd.in/deGU7E5t

FINALLY! 🔥 Mystic v2 in Magnific 🔥 Mystic v2 is the most advanced AI generator to date, now capable of outputting images in up to 4k resolution! 💥 Whether you’re aiming for ultra-realistic photos or dreamy watercolor effects, you can achieve it all with just a prompt and a few adjustments. So, what can we say? - We started with Upscaling. - Then came Transformation. - It was time to move into 🔥 Generation 🔥! We promised to be an AI suite. And we always deliver. It's in our DNA. Oh, and we’ve enhanced the user experience too, so every image now adapts seamlessly to your screen. It’s the small upgrades that make a big difference. 😉

🔥 Magnific Relight Update 🔥 This is ONLY one small step for Magnific, one giant leap for Product Placement. Now you can use the 'Preserve Details' toggle in Relight, so all the small details of your product remain untouched. This is just an example, and its parameters used, so you can see how powerful it is! Changed the background an lighting of a product using: ✍ Prompt: "A detailed watch in a wooden table, wooden table background, product placement" ✨ Life transfer strength: 60% 🎨 Style: "Just composition" And you can just change the prompt to get infinite scenarios! "A detailed watch in the jungle, volumetric light, beautiful", etc. Kudos to Karanva and Palma from Freepik that worked with us in these improvements. We are working on more things to keep improving Relight. We know it's still 'not there' yet. Stay tuned! 👀 Enjoy!

❤ Favs in Magnific ❤ Just a few days ago we told you about "Grid view", but the upgrades didn't stop there. We had to add "Favs", an easy and fast way to find your favorite upscales, relights and style transfer. Simply hit the heart in your images, and filter all your images easier than before. Drop a ❤ in the comments if you LOVE this as much as we do. Love is in the air!

🔥 Grid View in Magnific 🔥 Now you can go back in time and find your first upscales from prehistory, when dinosaurs roamed the Earth. It seems easy to make a grid view (and of course is not rocket science), but it wasn't that simple. Here's the explanation from the hacker himself, Emilio Nicolás Riquelme: 💬 "Well, in the end, I had to download more than 60 TB, load them into memory and then re-upload them transformed. And to avoid it taking a month to execute, I had to learn a new technology that was new to me and ended up having almost a thousand machines simultaneously doing the work. They knocked it out in 5 hours I know we could use dynamic transformation on demand with Akamai (for example) but I decided to do it this way for speed, I hope it's worth it! For transforming the 60TB of images, I used products from Google Cloud Platform: Google Cloud Task in conjunction with Google Cloud Run/Functions (they use both terms interchangeably for what sounds best)." ⚡ How many images do we have? We have around 10M unique jobs. Each job can have several images (the original, the transformed, the reference, the thumbnail, etc). So we have: Around 30M unique images. We had to get technical, we know some of you love these juicy tech details. 😜 But in case you're more of a visual person, here's what this all means 👇