Wan 2.2 does images too
2 megapixels, ~3 seconds, just $0.02 per image https://
replicate.com/prunaai/wan-2.
2-image
…
MULTIMODAL AI
-
Wan 2.2 Now Generates Images at 2 Megapixels for $0.02
By
–
-
What Caused AI Image Generation Leap Since March?
By
–
Any theories on what caused that leap since March? Likely to be new image generation?
-
AI trained on multiple human communication structures beyond code
By
–
There are many forms of structure humans have created to communicate meaning. Code is only one of them, but AI has trained on them all.
-
From VFX Dreams to AI Research: A Higher-Order Attack
By
–
Tried being a VFX artist for films. Flopped.
So, got into vision/ML research to build agents that did what I couldn’t.
Higher-order attack on the same dream. (eventually broadened my interest within vision/ML research, but that's how I got into AI research) -
Third-Person Video Game Style AI Visual of Woman in Paris
By
–
3. Vue à la troisième personne, style jeu vidéo, mannequin femme, marchant dans les rues de Paris. pic.twitter.com/yr5EZEVhue
— VISION IA (@vision_ia) 4 août 20253. Vue à la troisième personne, style jeu vidéo, mannequin femme, marchant dans les rues de Paris.
-
Veo 3 Generates Stunning Video Game Scenes in Real Locations
By
–
On peut demander à Veo 3 de créer des scènes de jeu vidéo dans des lieux célèbres.
— VISION IA (@vision_ia) 4 août 2025
Les résultats sont complètement fous.
Voici des exemples avec leurs prompts :
1. Vue à la troisième personne, femme, style jeu vidéo, marchant à l’intérieur du siège de Google à Mountain View,… pic.twitter.com/77CQxehcn3You can ask Veo 3 to create video game scenes in famous locations. The results are completely insane. Here are some examples with their prompts: 1. Third-person view, woman, video game style, walking inside Google's headquarters in Mountain View, California.
-
xAI Rapidly Launches Grok Models and Features in Under a Year
By
–
Aucun labo d’IA n’égale la vitesse de xAI avec une petite équipe en seulement deux ans. En moins d’un an, ils ont lancé :
– Grok 2 / mini
– Aurora – génération d’image par texte
– Bouton Grok – analyse de posts
– Application mobile
– Grok 3 + DeepSearch
– Mode vocal non censuré -
Blind Users Leverage AI for Privacy in Sensitive Personal Tasks
By
–
As a general purpose technology, AI has all kinds of unexpected uses. Example: this study finds blind users turn to AI to describe sensitive materials (pregnancy tests, checking appearance), realizing that it may not be 100% accurate, but it provides privacy where there was none
-
Reasoning Models Will Surpass Diffusion-Based Media Generation
By
–
Prediction: diffusion-based media models (images, video, music) are a dead end, and whoever cracks reasoning-based media models will win. We already saw a glimpse of this with OpenAI’s partially autoregressive native image generation – the level of control is far beyond what is
