Veo 3 img-to-video test, it seems to handle unusual animation styles pretty well/appropriately pic.twitter.com/mABBozFI6P
— fofr (@fofrAI) 9 juillet 2025
Veo 3 img-to-video test, it seems to handle unusual animation styles pretty well/appropriately
@fofrai
-
Veo 3 img-to-video test handles unusual animation styles
By
–
-
AI Voice Output Variations Based on Prompt Engineering
By
–
The voices all come out slightly differently depending on the prompt. In this one the accent also changed a fair bit toohttps://t.co/aUojReUPsL
— fofr (@fofrAI) 9 juillet 2025The voices all come out slightly differently depending on the prompt. In this one the accent also changed a fair bit too
-
Veo 3 Video AI: Scene Cutting and Character Consistency
By
–
> Immediately cut to a scene where she is walking through a summer park, she says to the camera as she walks: Scene ingredients don't work yet, but you can ask Veo 3 to cut to a new scene with the same character
— fofr (@fofrAI) 9 juillet 2025
Weirdly, when cutting to a new scene the following scene has that… pic.twitter.com/AgbYpHql8M> Immediately cut to a scene where she is walking through a summer park, she says to the camera as she walks: Scene ingredients don't work yet, but you can ask Veo 3 to cut to a new scene with the same character Weirdly, when cutting to a new scene the following scene has that
-
How AI Memes Perceive Their Own Existence Online
By
–
Asking memes how they feel about being memes pic.twitter.com/hdCuh7eigk
— fofr (@fofrAI) 9 juillet 2025Asking memes how they feel about being memes
-
Veo 3 reduces unwanted subtitles in video generation
By
–
Also new with Veo 3: – Reduction in unwanted subtitles (I wonder if this is a prompting trick in Flow)
-
LTX LoRAs for Generative AI Experimentation
By
–
One of those days where I'm chatting with @allnoteson and we're thinking how great it'd be if there were some neat LTX loras to experiment with.
— fofr (@fofrAI) 8 juillet 2025
Not even 1hr later… https://t.co/vQTq4mEZUUOne of those days where I'm chatting with @allnoteson and we're thinking how great it'd be if there were some neat LTX loras to experiment with. Not even 1hr later…
-
Black Forest Labs AI Model on Replicate Platform
By
–
Yep https://
replicate.com/black-forest-l
abs
… -
Minimax Hailuo-02 AI Model Now Available on Replicate
By
–
Here you gohttps://t.co/z8xyIA5TwG https://t.co/WorYHOko5o
— fofr (@fofrAI) 8 juillet 2025Here you go https://
replicate.com/minimax/hailuo
-02
… -
Comparing Recent AI Video Models on Replicate
By
–
I wrote a blog post comparing all the recent video models on Replicate https://
replicate.com/blog/compare-a
i-video-models
… -
AI System Reads Text on Sweater Using Vision Capabilities
By
–
Also, she can read out what's on her sweater, which is kind of neat.
— fofr (@fofrAI) 7 juillet 2025
> She reads out loud what it says on her sweater, while pointing to the words https://t.co/als4HMlLGP pic.twitter.com/s8fZvztSXsAlso, she can read out what's on her sweater, which is kind of neat. > She reads out loud what it says on her sweater, while pointing to the words