Show HN: Efficient LLM Architectures for 32GB RAM (Ternary and Sparse Inference)github.com·2 pts·fatihturker·1
Could ternary weights make 500B models runnable on consumer hardware?opengraviton.github.io·7 pts·fatihturker·6
Show HN: OpenGraviton – Run 500B+ parameter models on a consumer Mac Miniopengraviton.github.io·13 pts·fatihturker·5