Weight-only INT4 for RTX 3090/4090-class cards, where FP8 and NVFP4 are emulated or unusable. Full recipe on every card.
Alex Αdamopoulos
aleada
·
AI & ML interests
Finetuning, models quantization, agents, RAG, integration,
Recent Activity
updated a model 13 days ago
aleada/SmolLM3-3B-W4A16 updated a model 13 days ago
aleada/Pixtral-12B-W4A16 updated a model 13 days ago
aleada/Phi-4-mini-instruct-W4A16Organizations
None yet