Stable Diffusion 3.5 Large

8.1B parameters • 4.9GB VRAM (Q4 estimate) • 4,096 context

Estimate · Q4_K_M · 4k context · batch 1. For dense 7B–70B this sits ~5–8% above the Q4_K_M GGUF file. Not peak runtime VRAM and not a measured bench.