CREATIVE AI - AI Dynamics

Veo 3 generates video games as community theater productions

By

–

19 July 2025 5h07

" as a community theater production" may be one of the most delightful Veo 3 Fast prompts

Please enjoy, in order: GTA, Pokemon, Mario Kart, The Witcher 3, Stardew Valley, Tetris, Mortal Kombat, The Sims, & Death Stranding(!)

Yes, the whole prompt was the one above. pic.twitter.com/ylG8vbznZj
— Ethan Mollick (@emollick) 19 juillet 2025

" as a community theater production" may be one of the most delightful Veo 3 Fast prompts Please enjoy, in order: GTA, Pokemon, Mario Kart, The Witcher 3, Stardew Valley, Tetris, Mortal Kombat, The Sims, & Death Stranding(!) Yes, the whole prompt was the one above.

→ View original post on X — @emollick

19 July 2025

GPT-4.5 and Suno Integration Sparks Creative Interest

By

@petergostev

–

19 July 2025 0h49

Haha, I kind of like it though, gpt-4.5 + Suno

→ View original post on X — @petergostev

19 July 2025

DeepAgent Generates Production-Quality PowerPoints Instantly

By

@abacusai

–

18 July 2025 20h03

DeepAgent makes production quality powerpoints in one shot pic.twitter.com/emAeXGX3bW
— Abacus.AI (@abacusai) 18 juillet 2025

DeepAgent makes production quality powerpoints in one shot

→ View original post on X — @abacusai

18 July 2025

Using Otio to summarize YouTube videos with AI

By

@jouhatsu_ai

–

18 July 2025 20h00

4- Plus besoin de prendre des notes sur YouTube

Otio extrait les points clés des vidéos longues.

Plus besoin de prendre des notes.
Posez vos questions.

Gagnez du temps. Apprenez plus vite. pic.twitter.com/o4MQdE0fx5
— Jouhatsu | AI Influence Operator (@Jouhatsu_ai) 18 juillet 2025

4- Plus besoin de prendre des notes sur YouTube Otio extrait les points clés des vidéos longues. Plus besoin de prendre des notes.
Posez vos questions. Gagnez du temps. Apprenez plus vite.

→ Voir le post original sur X — @jouhatsu_ai

18 July 2025

Lipsync-2 Synchronizes Lips to Speech in Real Time

By

@replicate

–

18 July 2025 19h48

Lipsync-2 by @synclabs_so is now available on Replicate!

Lipsync 2.0 instantly syncs lips to speech with realistic motion and style—no training, just drop in audio.

Test it out: https://t.co/mKpzmZDgtg pic.twitter.com/H4140DqexF
— Replicate (@replicate) 18 juillet 2025

Lipsync-2 by @synclabs_so is now available on Replicate! Lipsync 2.0 instantly syncs lips to speech with realistic motion and style—no training, just drop in audio. Test it out: http://replicate.com/sync/lipsync-2

→ View original post on X — @replicate

18 July 2025

Human Creativity Remains Currency Amid AI Content Surge

By

@mustafasuleyman

–

18 July 2025 19h29

For all the fear about a deluge of AI-generated content, I genuinely believe that creativity will remain the real currency. Human ingenuity, style, craft are going to matter more not less.

→ View original post on X — @mustafasuleyman

18 July 2025

Runway Gen-4 Best Model for Identity Preservation Video

By

@fofrai

–

18 July 2025 16h03

A few folks have asked me which model is best at preserving identity. If I had to pick one, purely on character consistency, I'd go with Runway's Gen-4. https://
replicate.com/runwayml/gen4-
image
… Close up outputs are accurate and you can prompt many different scenes. Unlike kontext-pro you don't

→ View original post on X — @fofrai

18 July 2025

Hailuo 02 AI Generates Realistic Zoom Effect Video

By

@vision_ia

–

18 July 2025 16h00

Effet de zoom avec Hailuo 02. Cette vidéo n'est pas réelle. https://t.co/Zgo26XUldj pic.twitter.com/S5SC0cM1j3
— VISION IA (@vision_ia) 18 juillet 2025

Effet de zoom avec Hailuo 02. Cette vidéo n'est pas réelle.

→ Voir le post original sur X — @vision_ia

18 July 2025

Training and Running Fast Flux Models with LoRAs

By

@fofrai

–

18 July 2025 15h55

You can train and then run from here https://
replicate.com/replicate/fast
-flux-kontext-trainer/train
… Or do you mean loras you have already trained?

→ View original post on X — @fofrai

18 July 2025

Fine-tuned LoRA Model Transforms Faces Into Character Style

By

@fofrai

–

18 July 2025 15h27

I trained a kontext lora on my 0_1 webp character. The aim was to see if I could make it change any person in an input image into 0_1. https://
replicate.com/fofr/kontext-0
_1-webp
… It kind of works. It's very hit or miss. Forcing an aspect ratio change makes it more likely to work.

→ View original post on X — @fofrai

18 July 2025