The Local LLM Index / Quantization & Formats / #141
fkkarakurt/Nerve
by fkkarakurt · Quantization & Formats · updated 10d ago
Run a real LLM, sentence embeddings and on-device learning in pure C - one header, zero dependencies. TinyLlama-1.1B on a 2017 laptop CPU; the same code as a 65 KB WebAssembly module in your browser. Deterministic, auditable, embeddable anywhere C compiles.
50
momentum
52
stars
2
forks
#141
rank
ansi-cartificial-intelligencecc99edge-aiembedded-systemsembeddingsheader-onlyinference-enginellmlocal-aimachine-learning
View on GitHub →