maharshi's banner
maharshi's profile picture

maharshi

@maharshii • 42,714 subscribers

learning deeply about life one gradient step at a time - ml perf optimizer @ fal

Shorts

i made hunyuan-avatar faster: it can now generate upto 400 frames at 25 fps (16s video with audio) in only 5-6 minutes, it used to take ~15 minutes to generate a 5s video before!

i made hunyuan-avatar faster: it can now generate upto 400 frames at 25 fps (16s video with audio) in only 5-6 minutes, it used to take ~15 minutes to generate a 5s video before!

27,680 просмотров

Videos

LIVE
1.2k

Anya Rossi

sweetdream.ai

SweetDream.aiSponsored•Livecam
Streaming Now

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

HD live stream
Exclusive private shows
1.2k viewers online

Free preview available • Premium content

Elon Musk with interstellar’s music, it is heavenly.
0:22
maharshii's profile picture

Elon Musk with interstellar’s music, it is heavenly.

maharshi

654,254 просмотров • 1 год назад

as someone who dances i always wanted to do it below the japanese cherry blossoms, ah i miss this place a lot
0:24
maharshii's profile picture

as someone who dances i always wanted to do it below the japanese cherry blossoms, ah i miss this place a lot

maharshi

16,199 просмотров • 12 дней назад

holy freaking shit, who makes videos of this quality for free!? everyone go watch it on 2swap YT channel.
0:56
maharshii's profile picture

holy freaking shit, who makes videos of this quality for free!? everyone go watch it on 2swap YT channel.

maharshi

114,098 просмотров • 10 месяцев назад

the chorus in the background of this part of the song ALWAYS gets me, this is so peak man i can’t-
0:25
maharshii's profile picture

the chorus in the background of this part of the song ALWAYS gets me, this is so peak man i can’t-

maharshi

19,913 просмотров • 2 месяцев назад

i was playing around the newest TTS (text to speech) model 'kokoro' which has just 82M parameters! by adding only 1 line of code you can create a custom voice that mixes any two (out of ten voices) in any ratio example: mixing one male and one female voice in the 60-40 ratio
0:24
maharshii's profile picture

i was playing around the newest TTS (text to speech) model 'kokoro' which has just 82M parameters! by adding only 1 line of code you can create a custom voice that mixes any two (out of ten voices) in any ratio example: mixing one male and one female voice in the 60-40 ratio

maharshi

88,729 просмотров • 1 год назад

Больше нет контента для загрузки