The Fine-Tuning Index / Quantize & Merge / #96
IST-DASLab/GSQ
by IST-DASLab · Quantize & Merge · updated 3mo ago
Gumbel-Softmax post-training quantization for LLMs (1–3 bit scalar, INT/GGUF-compatible).
36
momentum
121
stars
20
forks
#96
rank
ggufllmpost-training-quantizationquantization
View on GitHub →