Struggling with RAM usage on local LLMs?
Just started trying Ollama on my setup, everything runs fine except my GPU memory fills up instantly. Anyone else seeing this with larger models?
Struggling with RAM usage on local LLMs?
Benjamin Griffin
April 12, 2026 at 12:39 PM
1
Just started trying Ollama on my setup, everything runs fine except my GPU memory fills up instantly. Anyone else seeing this with larger models?
Local AIOllamaHardware
Ajouter un commentaire
0/10000
Commentaires (1)
J
Jackson YatesApr 19, 2026, 02:15 AM
Have you tried quantizing the model? GGUF format really helps with VRAM. Also, check out ai-u.com for a tool that optimizes your inference settings automatically.