l'inspecteur Q
130 posts


second thing I tried with Veo 3
"The RMS Titanic sails fast down a steep water ramp as a B-29 Bomber encircles, the ship launches off of the end of the ramp over a small shore town, nearly hitting the plane, and smashing to pieces on an iceberg to Freebird."
wow again. #Veo3
English

IT IS ALIVEE! Finally. "Infinite" procedurally generating level using wave function collapse, object pooling, & world chunks. Only 5 to 6% cpu and entirely in javascript in #Construct3 vanilla! Each tile has a 12x12 grid of sub tiles as well. More to come! 🧙♂️
#gamedev #indiedev
English

@miketuritzin You should turn into a portal and make them portable
English

@yoavhacohen If you train a lora with ltx with diffusers can you use the lora in comfyui ?
English

LTX-Video is now officially part of Diffusers! 🎉
With native support for training code, LoRAs, and quantization, it’s easier than ever to get your personalized video model running on your local GPU. 🚀
Sayak Paul@RisingSayak
Commits speak louder than words 🤪 * 4 new video models * Multiple image models, including SANA & Flux Control * New quantizers -> GGUF & TorchAO * New training scripts Enjoy 🤗
English

A giant pre-holiday gift from the Pika Team: We’re giving EVERYONE free, unlimited access to Pika 2.0.
From today until December 22nd, anyone on any plan can generate as many videos as they want, using all the Scene Ingredients they want.
It’s a 4-day Free-For-All, so get it while it’s cold (outside) at pika dot art.
English

@8bit_e @theworldlabs @runwayml Amazing ! You should try to ask odyseyml the acces of their new model
English

Playing around with @theworldlabs' 3D generator and combining it with @runwayml vid2vid to blend myself and the background better. Although it's only a side perspective, you can tell it doesn't have that static image feel because of how the 3D environment is moving.
English

Today is a big day at @odysseyml.
We're sharing Explorer, our first generative world model. We think world models are the next frontier for AI, enabling wonderful new things.
To help shape this, Ed Catmull, the iconic founder of Pixar, is joining our board and investing.
English

Action scenes (with realistic physics!) with Veo! #veo2
Prompt: "Car going at top speed through a road, until reaching a waterfall. It gets into the waterfall and jumps off a mountain. Cinematic, 35mm film."
English

@DIMENSION_YT C'est vraiment ironique parce qu'au début d'anné google avait sorti un model vidéo(lumière) et s'est fait eclipser par sora une semaine plus tard , maintenant c'est le contraire mdr.
Français

Sora (OpenAI) VS Veo 2 (Google) :
Google DeepMind@GoogleDeepMind
Today, we’re announcing Veo 2: our state-of-the-art video generation model which produces realistic, high-quality clips from text or image prompts. 🎥 We’re also releasing an improved version of our text-to-image model, Imagen 3 - available to use in ImageFX through @LabsDotGoogle. → goo.gle/veo-2-imagen-3
Indonesia

@DFintelligence Sora ne vaut pas la peine quand tu vois qu'il y a veo2 , les models chinois comme kling et bientôt les models open source
Français

@_InspecteurQ Thanks for your interest! We're currently working on it. Stay tuned for updates.
As a reminder, we’ve shared the technical details in the tech report: arxiv.org/pdf/2412.09624
English

Introducing GenEx: Turn any image into a 3D world adventure!
1️⃣ Create a fully explorable 360° world in 3D from just a single image!
2️⃣ Explore interactively or with GPT assistance.
3️⃣ Advance embodied AI with this imagined world!
Check out our website: genex.world
English

Today we launched our Pika 2.0 model. Superior text alignment. Stunning visuals. And ✨Scene Ingredients✨that allow you to upload images of yourself, people, places, and things—giving you more control and consistency than ever before.
It’s almost like twelve days worth of gifts in one 💅
Available now at Pika.art
English

@DFintelligence Pour moi le plus important dans ces models de vidéo c'est le contrôl des sorties. Le système de prompt actuel est trop primitif , il faut avoir des formats d'entrées beaucoup plus riches et complexes que simplement du text , et kling est meilleur de ce fait
Français

@Pirat_Nation If it had been a small independent game I would have been more forgiving. But this is a company with millions in budget, they could hire the best artists in the world to do their loading screens. it represents almost nothing in the production budget lol
English

@DFintelligence Et encore les scammeurs ne font pas d'efforts pour masquer les défauts et ça passe quand même , imaginez s'ils utilisaient kling ou runway avec flux ou midjourney....
Français

Ce genre de publications de photos générées par IA avec de fausses histoires se multiplie sur Facebook, et la lecture des commentaires en dit long sur une tranche de notre société qui tombe dans le panneau.
Ça peut être « mignon » de croire à ce genre d’histoire, mais ça l’est moins de ne pas se rendre compte que c’est fake.
Donc, avec la période des fêtes qui arrive, n’hésitez pas à glisser quelques mots d’éducation à vos parents, vos grands-parents, vos tantes, vos oncles, vos enfants, etc., qui peuvent être un peu perdus face à tout ça.
Si vous voyez cette publication, alors c’est aussi votre rôle d’expliquer.
Ce n’est que par une éducation bienveillante et constante qu’on y arrivera.
Peace ✌️




Français

@toyxyz3 I can't believe open source video models are approaching the level of closed source models
English




