NativeTernary: A Self-Delimiting Binary Encoding with Unary Run-Length Hierarchy Markers for Ternary Neural Network Weights, Structured Data, and General Computing Infrastructure
Not on the map. Why?
Abstract (arXiv)
BitNet b1.58 (Ma et al., 2024) demonstrates that large language models can operate entirely on ternary weights {-1, 0, +1}, yet no native binary wire format exists for such models. NativeTernary closes this gap. Benchmarked against GGUF on the real BitNet b1.58 2B4T architecture (24 layers, ~170 tensors, 2B parameters): NativeTernary encodes ternary weights at exactly 2.000 bits per weight -- 1.31x smaller than GGUF Q2_K and 4.0x smaller than GGUF int8 -- while reducing boundary and framing overhead by 460x (91 bytes vs ~42KB of GGUF tensor headers). Encode throughput: 47--69 MB/s. Decode throughput: 35--45 MB/s on commodity hardware. The decoder is a 10-line stateless state machine resilient to bitstream corruption.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 2 of its 8 references that are in the map. First posted on arXiv 3 Apr 2026.
Builds on
2 of 8 references.
References · 2
- 2020Longformer: The Long-Document Transformercited by 3,413
- 2024The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bitscited by 308
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025BitNet b1.58 2B4T Technical Report
- 2026Breaking the 1.58-bit Barrier for Ternary LLMs
- 2024When are 1.58 bits enough? A Bottom-up Exploration of BitNet Quantization
- 2026Lost in the bf16 Cast: Exporting Ternary Language Models Can Revert Most Low-Learning-Rate Code Changes
- 2023BitNet: Scaling 1-bit Transformers for Large Language Models
- 2026Capability-Stratified Degradation in Ternary Language Models
- 2023Binary and Ternary Natural Language Generation
- 2025Bitnet.cpp: Efficient Edge Inference for Ternary LLMs