No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from prism-ml/Ternary-Bonsai-4B-gguf
All models by prism-ml →
Prism ML Website | White Paper | Demo & Examples | Discord
Ternary (1.58-bit) language model in GGUF Q2_0 format for llama.cpp
llama.cpp. Use our fork at PrismML-Eng/llama.cpp (prism branch, default) which adds Q2_0 support for CPU (NEON/generic) and Metal. Upstream PR coming soon.Each weight takes a value from {-1, 0, +1}, with one shared FP16 scale per group of 128 weights:
Q2_0 encodes each weight as a 2-bit code q in {0, 1, 2, 3}, dequantized via w = (q - 1) * scale. One 128-element block is 34 bytes (2 bytes FP16 scale + 32 bytes of packed 2-bit codes) for an effective 2.125 bits/weight. The fourth code point (q = 3, reconstructing to +2 * scale) is reserved for future extensions; for ternary weights it is unused.
llama.cpp CLIllama.cpp serverFlags: -ngl 99 -fa 1 for Metal; -ngl 0 -fa 1 -t 10 for CPU.
Q2_0 is effectively lossless for ternary weights — the ternary values land exactly on three of the four 2-bit code points, so quantize/dequantize is bit-exact in the absence of FP16 scale rounding.
Evaluated with EvalScope v1.4.2 + vLLM 0.15.1 on NVIDIA H100. Full benchmark suite:
For questions, feedback, or collaboration inquiries: contact@prismml.com
w_i = scale_g * t_i, t_i in {-1, 0, +1}git clone https://github.com/PrismML-Eng/llama.cppcd llama.cppcmake -B build -DGGML_METAL=ON # or -DGGML_CUDA=ON, -DGGML_VULKAN=ONcmake --build build -j./build/bin/llama-cli \ -m Ternary-Bonsai-4B-Q2_0.gguf \ -p "Explain quantum computing in simple terms." \ -n 256./build/bin/llama-server -m Ternary-Bonsai-4B-Q2_0.gguf -c 4096density = -ln(1 - score/100) / size_GB@techreport{ternarybonsai, title = {Ternary Bonsai: 1.58-bit Language Models at 8B, 4B, and 1.7B Scale}, author = {Prism ML}, year = {2026}, month = {April}, url = {https://prismml.com}}