logoalt Hacker News

2001zhaozhaoyesterday at 10:32 PM1 replyview on HN

I think if they made this for Qwen3.8-Next it could fit in a single 5090?


Replies

kennywinkertoday at 5:27 AM

180b * 1.76 bits per weight = 39.6 gigabytes.

Best you could realistically run in 32gb is like 28gb, or a 127B param model