Yoshinari Fujinuma
@akkikiki.bsky.social
Member of Technical Staff@Cantina Labs (NYC);
CS PhD @CUBoulder;
Posts are my own;
Website: https://akkikiki.github.io/
Substack: http://substack.com/@akkikiki
Lived: 🇹🇭🇯🇵🇫🇷🇺🇸
Posts: JA/EN
CS PhD @CUBoulder;
Posts are my own;
Website: https://akkikiki.github.io/
Substack: http://substack.com/@akkikiki
Lived: 🇹🇭🇯🇵🇫🇷🇺🇸
Posts: JA/EN
I played around with the mlx and 4-bit version of Qwen3-30B-A3B locally on an Apple M4 Max chip, with Japanese and English. This is amazing. It seems feasible to run it locally for tasks other than long-horizon or complex ones.
April 30, 2025 at 9:42 PM
I played around with the mlx and 4-bit version of Qwen3-30B-A3B locally on an Apple M4 Max chip, with Japanese and English. This is amazing. It seems feasible to run it locally for tasks other than long-horizon or complex ones.