PixVerse v4 takes 23s for 540p, it's fast, smooth and high quality. Outputs can have an AI look to them https://
replicate.com/pixverse/pixve
rse-v4
… Ray Flash 2 takes 30s: https://
replicate.com/luma/ray-flash
-2-540p
… Wan 2.1 480p takes 30s, more realism but lower frame rate https://
replicate.com/wavespeedai/wa
n-2.1-t2v-480p
… They're all listed
@fofrai
-
PixVerse v4, Ray Flash 2, Wan 2.1: AI video generation tools compared
By
–
-
PixVerse v4 Excels at TikTok Dance Video Generation
By
–
PixVerse v4 is generally pretty good at the tiktok dance style, especially if you put it into smooth mode.
— fofr (@fofrAI) 8 mai 2025
> a tiktok dancer
9:16, smooth, 720p, 5shttps://t.co/lGOzt5UGzf pic.twitter.com/LecWzkkQc5PixVerse v4 is generally pretty good at the tiktok dance style, especially if you put it into smooth mode. > a tiktok dancer 9:16, smooth, 720p, 5s https://
replicate.com/p/n7e6n2031nrm
80cpnjw9jr75mr
… -
PixVerse v4 video generation model passes TikTok dancer test
By
–
PixVerse v4 passes the "tiktok dancer on drone" test, albeit with a bit less coherency than Wan 2.1 (but it makes up for this with frame rate)https://t.co/QsZQhLtzfA https://t.co/RhiXYlG4st pic.twitter.com/ykId1gXRIE
— fofr (@fofrAI) 7 mai 2025PixVerse v4 passes the "tiktok dancer on drone" test, albeit with a bit less coherency than Wan 2.1 (but it makes up for this with frame rate) https://
replicate.com/p/6p4ypb98csrm
80cpnk59mwyn14
… -
PixVerse v4 Image Generation Tool Launches on Replicate Platform
By
–
Ideogram v3 prompt: > a closeup portrait photo of an anime inspired flower-punk woman, bokeh and beautiful light, over the photo is the text "PixVerse v4" and the subtitle "on Replicate", the text is center middle in a serif typeface, there is pink and gold, the text is white
-
Custom ASMR Voice Trained for Minimax Text-to-Speech
By
–
I wrote a blog post. I also trained an ASMR voice for Minimax speech-02, and then I had it read the blog for me.
— fofr (@fofrAI) 6 mai 2025
The voice_id to use with the speech-02 models is R8_FDU1SV5Shttps://t.co/ZI8d1XMX1k https://t.co/BqCsqiTkNo pic.twitter.com/S0ObYoAZq1I wrote a blog post. I also trained an ASMR voice for Minimax speech-02, and then I had it read the blog for me. The voice_id to use with the speech-02 models is R8_FDU1SV5S https://
replicate.com/blog/minimax-t
ext-to-speech
… -
GPT-4o Demonstrates Advanced Image Understanding Capabilities
By
–
4o can do this: > a photo of a man wearing two collared t-shirts
-
Ideogram’s Celebrity Image Generation: Resemblance Without Accuracy
By
–
Ideogram seems to have some knowledge of celebrities, but weirdly their images do look like them, but as if someone different was trying to play them in a movie. mark zuckerberg, david duchovny, emma watson, chiwetel ejiofor
-
Ideogram v3 Creates Artistic Text with Monstera Leaf Patterns
By
–
Ideogram v3 again: > in swirling free-flowing words "Enjoy a slice of fofr's perfectly melted ai" written in beautiful stretched lettering with interesting interconnections and glyphs and ligatures, fills the space, repeating pattern of monstera leaves, all the letters
-
Cyclist Intentionally Interferes with Waymo Autonomous Vehicle
By
–
I saw a cyclist do this to a Waymo intentionally https://t.co/YsEboSphdt
— fofr (@fofrAI) 4 mai 2025I saw a cyclist do this to a Waymo intentionally
-
Ideogram v3 generates artistic text design with monstera pattern
By
–
Ideogram v3 again: > in swirling free-flowing words "Enjoy a slice of fofr's perfectly melted ai" written in beautiful stretched lettering with interesting interconnections and glyphs and ligatures, fills the space, repeating pattern of monstera leaves, all the letters