GPU × Model compatibility
AIPCFit compares the model's recorded GGUF variants against the12GB of dedicated VRAM on the RTX 3060 12GB. The result is a deterministic memory-planning check, not a speed benchmark.
Quick answer
At least one recorded variant reaches AIPCFit's good-fit or fits threshold using this GPU's dedicated VRAM. Exact speed, context capacity, and runtime behavior are not predicted here.
Variant-by-variant fit
GGUF · Q4_K_M
12 GB dedicated VRAM meets the 4.2 GB planning memory level for Phi-3.5 Mini · Q4_K_M.
This is a deterministic memory fit, not a speed or quality benchmark.
AIPCFit does not invent tokens-per-second numbers.
This page does not assume how much RAM your full PC has, so partial offload is not ranked.
Longer context and runtime settings can increase real memory pressure.