Nanotern – Ternary LLM training on consumer hardware
Comments
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Summary
Nanotern is an open-source project that trained a 126M-parameter transformer three ways on a single consumer RTX 4090: full bf16, naive post-training ternarization, and quantization-aware training with ternary weights from random init.