We've had so much fun making these https://
replicate.com/flux-kontext-a
pps
… These are outputs from the "portrait-series" model, it's like my old consistent character model, but it's cheaper and faster, and also, wow. (Top left is the original)
MULTIMODAL AI
-

Flux Kontext Portrait Series Model for Faster Cheaper Image Generation
By
–
-

AI Reimagines Vase in Colorful Aurora Style
By
–
This was one of my favourites > A photo of the reimagined vase in the colorful aurora style
-

Kontext Multi-Image AI Tool Enables Advanced Image Editing
By
–
Here's an experimental version of Kontext, powered by [max], that can take two images: https://
replicate.com/flux-kontext-a
pps/multi-image-kontext
… Prompt:
> Put the woman into a white t-shirt with the text on it -

AI-Powered Image Editing and Generation with Photoshop and OpenAI
By
–



Wow ! Photoshop et OpenAI en PLS. Incroyable Kontext de @bfl_ml
, bravo !! Vous pouvez modifier tout ce que vous voulez sur une image, la transformer dans le style de votre choix, ou utiliser n'importe quel style pour générer de nouvelles images. J'ai voulu tester avec des -

Change Someone’s Hair Style or Hair Color Online
By
–
Change someone's hair style or hair color: https://
replicate.com/flux-kontext-a
pps/change-haircut
… -

Put Yourself into Iconic Locations with Flux Kontext
By
–
Put yourself into an iconic location https://
replicate.com/flux-kontext-a
pps/iconic-locations
… -
Google AI Lets Single User Generate Full Movie Trailer
By
–
Une bande-annonce entière générée par un seul utilisateur avec l’IA de Google.
— VISION IA (@vision_ia) 29 mai 2025
Et ce n’est que la version la plus mauvaise que nous n’aurons jamais de cette technologie.
Lentement, Hollywood devient un plug-in.
Dans 2 ans, on regarde quoi ? pic.twitter.com/Z7Bq7nEIOpUne bande-annonce entière générée par un seul utilisateur avec l’IA de Google. Et ce n’est que la version la plus mauvaise que nous n’aurons jamais de cette technologie. Lentement, Hollywood devient un plug-in. Dans 2 ans, on regarde quoi ?
-
Hume unveils EVI 3 voice model
By
–
Hume released EVI 3 🔥
— 🚨 AI News | TestingCatalog (@testingcatalog) 29 mai 2025
"EVI 3 is a speech-language model that can understand and generate any human voice" https://t.co/nM30Qc6umx pic.twitter.com/q0i0JUi0mvHume released EVI 3 "EVI 3 is a speech-language model that can understand and generate any human voice"
-

Flux Model Combines Realism with Customization Capabilities
By
–
New Flux model looks crazy good. Imagine the realism of Flux combined with the editability and customization of GPT-4o. Super promising.
-
Deep Research and Labs: Advanced AI Tools for Comprehensive Analysis
By
–
Deep Research remains the fastest way to get comprehensive answers to in-depth questions.
— Perplexity (@perplexity_ai) 29 mai 2025
Labs is designed to invest more time and leverage multiple tools, such as coding, headless browsing, and design to create more dynamic outputs. pic.twitter.com/DcsFBmladXDeep Research remains the fastest way to get comprehensive answers to in-depth questions. Labs is designed to invest more time and leverage multiple tools, such as coding, headless browsing, and design to create more dynamic outputs.