Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Mistral NeMo. Parent comment
- 3B for CPU inference or running on edge devices.
- 20-30B for maximizing single consumer GPU potential.
- 70B+ for those who can afford it.
7-9B never felt like an ideal size.
There are no comments on this comment yet.
- 3B for CPU inference or running on edge devices.
- 20-30B for maximizing single consumer GPU potential.
- 70B+ for those who can afford it.
7-9B never felt like an ideal size.