Show HN: QKV Core – Run 7B LLMs on 4GB VRAM via surgical memory alignment

1 pointsposted 2 months ago
by broxytr
(github.com)

No comments yet