Datasets:

dispatchAI
/

quantization-guide

level stringclasses 7 values	size_vs_fp16 stringclasses 7 values	quality stringclasses 7 values	use_case stringclasses 7 values	recommended bool 2 classes
Q2_K	25%	75%	Ultra-low RAM (1GB devices)	false
Q3_K_M	30%	85%	Very constrained devices	false
Q4_K_M	40%	92%	Sweet spot for mobile	true
Q5_K_M	50%	95%	Quality-sensitive mobile	true
Q6_K	60%	97%	Near-lossless mobile	false
Q8_0	70%	98%	High-quality, smaller than FP16	false
F16	100%	100%	Reference / debugging	false

Quantization Guide

Reference for choosing the right GGUF quantization level for mobile deployment.

Q4_K_M is the recommended sweet spot — 40% of FP16 size, 92% quality.