FitMyModel

What local AI models run on 12 GB of VRAM?

Last updated

We earn from qualifying purchases via affiliate links. Disclosure.

2 of 13 ranked open models fit in 12 GB: 1 with 4 GB or more to spare and 1 tight. The best ranked is Phi-4 14B (9.1 GB).

ModelLMArenaNeedsFitPages
Phi-4 14BMicrosoft · Q4_K_M
 #322 9.1 GB Tight Arc B580 · Guide · 8–12 GB GPU tier
Llama 3.1 8BMeta · Q4_K_M
 #345 4.9 GB Fits Arc B580 · Guide · 8–12 GB GPU tier

Fit compares the Ollama library download size with 12 GB. "Fits" means 4 GB or more spare and "Tight" means under 4 GB; both bands are our estimate, and tight models need a short context. Ranks from LMArena Text, checked Oct 27, 2026, 2:16 PM CT. No speeds are listed because we have no sourced numbers.

Hardware with 12 GB for the GPU

Prices: Newegg checked Oct 27, 2026, about 12:35 PM CT; Minisforum checked Oct 27, 2026, 12:40 PM CT. New condition only.

Other memory sizes

← 8 GB 8 GB 12 GB 16 GB 24 GB 32 GB 48 GB 64 GB 96 GB 128 GB 16 GB →

Try the What fits my memory tool or browse every pairing on Can it run?

FAQ

What local AI models run on 12 GB of VRAM?

2 of the 13 leaderboard-ranked open models we track fit in 12 GB: Phi-4 14B (9.1 GB), Llama 3.1 8B (4.9 GB). Sizes are Ollama library downloads; Phi-4 14B is tight (under 4 GB spare, our estimate).

What is the best-ranked local model for 12 GB?

Phi-4 14B, #322 overall on LMArena Text (checked Oct 27, 2026, 2:16 PM CT), needs 9.1 GB.

What is the largest model that fits in 12 GB?

Phi-4 14B at 9.1 GB (Q4_K_M). The runtime and context need roughly 1–4 GB more, which is our estimate.

Does 12 GB of unified memory count as 12 GB of VRAM?

No. On a Ryzen AI Max+ 395 system AMD says the GPU can use up to 48 GB of a 64 GB system and up to 96 GB of a 128 GB system, so use the GPU share when you read this page.