What hardware runs Phi-3.5-mini?

The cheapest cataloged machine that runs Phi-3.5-mini on the GPU is a AMD Radeon RX 7600 · 8GB VRAM. Here are the most affordable rigs that fit it, smallest memory first.

machinebest fitverdictspeed
AMD Radeon RX 7600 · 8GB VRAM 3.8B Q5_K_M ✅ Runs (tight) 19–30 tok/s
NVIDIA GeForce RTX 3070 · 8GB VRAM 3.8B Q5_K_M ✅ Runs (tight) 30–50 tok/s
NVIDIA GeForce RTX 4060 · 8GB VRAM 3.8B Q5_K_M ✅ Runs (tight) 19–30 tok/s
NVIDIA GeForce RTX 3080 · 10GB VRAM 3.8B Q8_0 ✅ Runs (tight) 45–70 tok/s
AMD Radeon RX 6700 XT · 12GB VRAM 3.8B Q8_0 ✅ Runs comfortably 20–35 tok/s
AMD Radeon RX 7700 XT · 12GB VRAM 3.8B Q8_0 ✅ Runs comfortably 25–40 tok/s
NVIDIA GeForce RTX 3060 · 12GB VRAM 3.8B Q8_0 ✅ Runs comfortably 20–35 tok/s
NVIDIA GeForce RTX 4070 · 12GB VRAM 3.8B Q8_0 ✅ Runs comfortably 30–50 tok/s
AMD Radeon RX 6800 XT · 16GB VRAM 3.8B Q8_0 ✅ Runs comfortably 30–50 tok/s
AMD Radeon RX 6950 XT · 16GB VRAM 3.8B Q8_0 ✅ Runs comfortably 35–55 tok/s
AMD Radeon RX 7800 XT · 16GB VRAM 3.8B Q8_0 ✅ Runs comfortably 35–60 tok/s
NVIDIA GeForce RTX 4080 · 16GB VRAM 3.8B Q8_0 ✅ Runs comfortably 45–70 tok/s

Already have a machine? See Phi-3.5-mini 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 phi-3.5-mini --hardware <your-rig>

Verdicts use the same GQA-aware fit math as the CLI — a AMD Radeon RX 7600 verdict here matches what runlocal tells you on the machine.