MoE models recomended for 8GB of VRAM + MoE offloading
Patryk
WhiskyAKM
AI & ML interests
LLMs for low end hardware
Recent Activity
updated a collection about 22 hours ago
Low Vram NVFP4 2026.09 updated a collection about 22 hours ago
Low Vram 2026.09 updated a collection about 22 hours ago
Low Vram NVFP4 2026.09Organizations
None yet
models 25
WhiskyAKM/Spark-X2.5-4B-NVFP4-GGUF
Text Generation • 4B • Updated • 144
WhiskyAKM/Spark-X2.5-4B-GGUF
Text Generation • 4B • Updated • 178
WhiskyAKM/Gemma-4-E2B-it-qat-GGUF
Any-to-Any • 5B • Updated • 1.13k
WhiskyAKM/MiniCPM5-2B-GGUF
Text Generation • 3B • Updated • 1.18k • 1
WhiskyAKM/MiniCPM5-2B-NVFP4-GGUF
Text Generation • 3B • Updated • 800 • 1
WhiskyAKM/Nemotron-3.5-Lightning-30B-A3B-GGUF
Text Generation • 33B • Updated • 1.71k
WhiskyAKM/Nemotron-3.5-Lightning-30B-A3B-NVFP4-GGUF
Text Generation • 33B • Updated • 612
WhiskyAKM/Ling-3.0-flash-GGUF
Text Generation • 127B • Updated • 534 • 4
WhiskyAKM/Ling-3.0-Tiny-NVFP4-GGUF
Text Generation • 8B • Updated • 334
WhiskyAKM/Ling-3.0-Tiny-GGUF
Text Generation • 8B • Updated • 440 • 3
datasets 0
None public yet