Friday, March 31, 2023

New best story on Hacker News: Llama.cpp 30B runs with only 6GB of RAM now

Llama.cpp 30B runs with only 6GB of RAM now
485 by msoad | 137 comments on Hacker News.


New best story on Hacker News: Astra can read and write weights to improve Qwen

Astra can read and write weights to improve Qwen 708 by dbreunig | 211 comments on Hacker News.