Show HN: 3.125-Bit LLM quantization bypassing tensor cores(blog.djellalmohamedaniss.workers.dev)
3 pointsby dmaniss2 hours ago1 comment