I wrote a blog post comparing all the recent video models on Replicate https://
replicate.com/blog/compare-a
i-video-models
…
CREATIVE AI
-
Comparing Recent AI Video Models on Replicate
By
–
-
Comprehensive Guide to Comparing AI Video Models
By
–
It's hard keeping up with all these new video models. We've put together a blog post to help you compare them.https://t.co/Ep89yJ40ER
— Replicate (@replicate) 8 juillet 2025
It covers:
– key specs like price, resolution, duration, fps, speed, and date of release
– features like text-to-video, image-to-video,… pic.twitter.com/hVqTx4YoyyIt's hard keeping up with all these new video models. We've put together a blog post to help you compare them. https://
replicate.com/blog/compare-a
i-video-models
… It covers: – key specs like price, resolution, duration, fps, speed, and date of release
– features like text-to-video, image-to-video, -
Hailuo 02 Model Now Available on Replicate Platform
By
–
Hailuo 02 is now live on Replicatehttps://t.co/tpHrS5ojPx
— Replicate (@replicate) 8 juillet 2025
Prompt:
"a dog speed climbs up a climbing wall at the olympics" pic.twitter.com/eTmNKWRSubHailuo 02 is now live on Replicate https://
replicate.com/minimax/hailuo
-02
… Prompt:
"a dog speed climbs up a climbing wall at the olympics" -
Guide: Creating First-Frame to Last-Frame Videos with Kontext
By
–
Here's my short guide on making good first-frame to last-frame videos using kontext
-
Steps for Creating First-to-Last Frame Videos with AI
By
–
Some helpful steps to make good first-to-last frame videos: 1. generate a start or end frame, we used seedream and asked for a woman holding up a poster with some text
2. use flux kontext to edit that frame, "remove the poster, put her hands down"
3. then we ran kontext again, -
Midjourney Video Creates Stunning Clay Animation Styles
By
–
Nouveaux styles pâte à modeler avec Midjourney Video.
— VISION IA (@vision_ia) 8 juillet 2025
Le résultat ? On dirait de la triche.
Voici des résultats à ne pas manquer 👇 pic.twitter.com/FMcGfUjrOsNouveaux styles pâte à modeler avec Midjourney Video. Le résultat ? On dirait de la triche. Voici des résultats à ne pas manquer
-
Veo 3 Animates Midjourney Images with Impressive Results
By
–
More strange cities, this time in Veo 3.
— Ethan Mollick (@emollick) 8 juillet 2025
I am pretty impressed by how well Veo 3 animates Midjourney images. All of these clips, including the sounds, were generated in a single shot from a Midjourney image & a simple prompt (I picked the best videos from a set of two options). pic.twitter.com/GKgJWZ0NLPMore strange cities, this time in Veo 3. I am pretty impressed by how well Veo 3 animates Midjourney images. All of these clips, including the sounds, were generated in a single shot from a Midjourney image & a simple prompt (I picked the best videos from a set of two options).
-
Frame to Video Feature Added to Flow Menu
By
–
Para usarlo sólo tenéis que seleccionar la opción Fotograma a Vídeo dentro del menú de Flow.
-
Veo 3 AI Model Interprets Talk Prompt Creatively
By
–
De hecho ha ocurrido una casualidad muy hermosa y es que le he subido una imagen mía de una charla con el prompt "drone shot of a man talking to big audience" y Veo 3 ha decidido que en la charla debía estar hablando de IA, vaya crack! 🔥 pic.twitter.com/uOVWtOQi9q
— Carlos Santana (@DotCSV) 8 juillet 2025De hecho ha ocurrido una casualidad muy hermosa y es que le he subido una imagen mía de una charla con el prompt "drone shot of a man talking to big audience" y Veo 3 ha decidido que en la charla debía estar hablando de IA, vaya crack!
-
VEO 3 Enhances Video Control with Image Frame Integration
By
–
🔴 ¡VIDEOS CON MÁS CONTROL en VEO 3!
— Carlos Santana (@DotCSV) 8 juillet 2025
El rey de la pista se actualiza en su plataforma Flow para aceptar el usar imágenes como fotograma inicial. Esto ofrece un control mucho más fino y desbloquea la posibilidad de crear tomas con consistencia de personajes, escenarios, etc! pic.twitter.com/nqPdcoWKvD¡VIDEOS CON MÁS CONTROL en VEO 3! El rey de la pista se actualiza en su plataforma Flow para aceptar el usar imágenes como fotograma inicial. Esto ofrece un control mucho más fino y desbloquea la posibilidad de crear tomas con consistencia de personajes, escenarios, etc!