What hardware runs Llama 3.2 Vision?

The cheapest cataloged machine that runs Llama 3.2 Vision on the GPU is a NVIDIA GeForce RTX 3080 · 10GB VRAM. Here are the most affordable rigs that fit it, smallest memory first.

machinebest fitverdictspeed
NVIDIA GeForce RTX 3080 · 10GB VRAM 11B Q4_K_M ✅ Runs (tight) 40–65 tok/s
AMD Radeon RX 6700 XT · 12GB VRAM 11B Q4_K_M ✅ Runs comfortably 20–35 tok/s
AMD Radeon RX 7700 XT · 12GB VRAM 11B Q4_K_M ✅ Runs comfortably 25–40 tok/s
NVIDIA GeForce RTX 3060 · 12GB VRAM 11B Q4_K_M ✅ Runs comfortably 20–35 tok/s
NVIDIA GeForce RTX 4070 · 12GB VRAM 11B Q4_K_M ✅ Runs comfortably 30–45 tok/s
AMD Radeon RX 6800 XT · 16GB VRAM 11B Q8_0 ✅ Runs (tight) 16–25 tok/s
AMD Radeon RX 6950 XT · 16GB VRAM 11B Q8_0 ✅ Runs (tight) 18–30 tok/s
AMD Radeon RX 7800 XT · 16GB VRAM 11B Q8_0 ✅ Runs (tight) 19–30 tok/s
Apple M1 · 16GB 11B Q4_K_M ✅ Runs (tight) 5–9 tok/s
Apple M1 Pro · 16GB 11B Q4_K_M ✅ Runs (tight) 14–25 tok/s
Apple M2 · 16GB 11B Q4_K_M ✅ Runs (tight) 8–13 tok/s
Apple M2 Pro · 16GB 11B Q4_K_M ✅ Runs (tight) 14–25 tok/s

Already have a machine? See Llama 3.2 Vision across popular rigs → or check your exact rig →

Planning a whole workflow? Rigs for a coding agent · local chat · reasoning · local RAG setup.

$npx runlocal-sh can-i-run llama-3.2-vision --hardware <your-rig>

Verdicts use the same GQA-aware fit math as the CLI — a NVIDIA GeForce RTX 3080 verdict here matches what runlocal tells you on the machine.