Parth Parmar

built akshara, a bpe tokenizer tuned for languages at 200k vocab vs gpt-4o's 0200k: - overall 5.03 vs 4.64 bytes/token - beats it on all 19 languages tested (wiki) - biggest gains: thai +52%, japanese +50%, bengali +32%, tamil +27% - code: ahead on 20 of 21, small gains - english ~tie (+1%) compression only, not Im quality akshara-tokenizer.vercel.app

updated my X profile

the non-commoditized part of the game

Garba Night day 1 1448 cals💪🏻

I try to make it with Claude Opus 5.5 in After Effects with higgsfield

@ChatGPT ate all frontier AI providers today and provided a one-step solution in one place.

good thing!

you can find in the menubar

pov: you are using secondary monitor with your mac and screen sharing is nightmare to change for multiple screens/windows.. building an app which will solve it!