Breaking the 1.58-bit Barrier for Ternary LLMs

Computer Science > Artificial Intelligence [Submitted on 14 Sep 2026] Title:Breaking the 1.58-bit Barrier for Ternary LLMs View PDF HTML (experimental)Abstract:Ternary Large Language Models (LLM) store every weight as one of three symbols $\{-1,0,+1\}$, so the cost of a ternary model is conventionally referenced to the information-theoretic $\log_2 3 \approx 1.585$ bits per weight. The prevailing ...