Image created with Flux Pro v1.1 Ultra. Image prompt: Assembly instruction diagram for a twin-lens reflex camera with film mechanism, vintage 1960s style, black leather and chrome colors, aged instruction manual paper, “IMAGES” in classic camera font, shutter mechanism detailed, film loading sequence shown
ByteDance’s Bagel 14B MOE (7B active) Multimodal with image generation (open source, apache license) is just an incredible modle. A unified multimodal model rivalling GPT-4o and Gemini 2.0, with 7B active params (14B total), 40K context, 88% GenEval and 85% understanding, https://x.com/rohanpaul_ai/status/1927705853580509607
Black Forest Labs – Frontier AI Lab https://bfl.ai/announcements/flux-1-kontext
There is something interesting in AI generated photos of simulated mundanity.”” / X https://x.com/emollick/status/1927512928313319573
Flux Kontext is out and it’s amazing! watch me build a Claude 4 enhanced image editor workflow on my iPhone in glif in 66 seconds https://x.com/fabianstelzer/status/1928433180765306968
Yesterday my brother came by. He asked for advice on how to use ai to create consistent technical drawings. I built a mini tool with @lovable_dev in 60 minutes for him. He was BLOWN away. This will save an employee 1,5 hours per day. Now he is making a list for more mini https://x.com/kattrisen/status/1922642975445627300
Insane. Nate Herkelman built a faceless Shorts machine for $0.75/video using AI tools in @n8n_io! 🤯 End-to-end automation that: ↳ generates close-ups + scenes ↳ selects best images ↳ auto-renders video ↳ posts to TikTok, IG, YT ↳ logs steps to GSheets Full demo in 🧵 ↓ https://x.com/DataChaz/status/1921981830594486383
OmniSync: Towards Universal Lip Synchronization via Diffusion Transformers https://ziqiaopeng.github.io/OmniSync/
Welcome to the Image Arena: FLUX.1 Kontext! 🖼️ 🎨 You’ll find that FLUX.1 Kontext Pro can generate AND edit images. Congrats to @bfl_ml on this exciting release. 🌲👏🌲 Check it out in the Arena, and get voting! https://x.com/lmarena_ai/status/1928236637709865217
I built an app with Lovable in one afternoon. It’s called Reflekt. You upload a photo. It breaks down the style, color, and layout. Then spits out a clean AI image style prompt. Try it here & let me know how it works for you: https://x.com/_Vikki_B/status/1921294175242219946
Benefits of BAGEL: – It’s one model that covers reading, reasoning, drawing, and editing without a quality bottleneck. – Supports long, mixed contexts: docs, tutorials, multi-image stories. – Works with arbitrary aspect ratios and multiple languages out of the box. – Plus, open https://x.com/TheTuringPost/status/1927123416823251389
A new recipe for training multimodal models 👉 Mixed together various data types: text next to images, video frames after captions, then webpages, etc. This way the model learns to connect what it reads with what it sees. ByteDance proposed and implemented this idea in their https://x.com/TheTuringPost/status/1927123359969468420
Remember the Great Ghiblification? Turns out, this is all part of a grander plan by OpenAI to make them appear cool and to win market share. Search, DeepResearch, Agents and Personlization with System Prompts, Tasks and the coming model unification are also part of that plan. https://x.com/scaling01/status/1926801814973804712
