benches/dgx-spark at main in okami.mom/llama-cpp-bonsai

skip to content

okami.mom / llama-cpp-bonsai

okami.mom/llama-cpp-bonsai

llama.cpp fork for Ternary Bonsai 2 27B: MTP speculation, suffix cache reuse (SCR), q8 KV, two-slot fairness, fused DeltaNet kernels

files main

/

languages

C++ 63%
C 14%
Python 8%
TypeScript 5%
GLSL 3%
Svelte 3%
other 4%

refs 2

main 8906fd8 Oct 18 bolt 87e1b42 Oct 16

≡ llama-cpp-bonsai

readme tickets …commits pipelines settings

dgx-spark/

dgx-spark/

benches/dgx-spark

benches/dgx-spark/ has no readme, its files are in the sidebar.