can-i-run-this-llm

Can I run this LLM?

Can a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) run Llama 3.3 70B Instruct?

Llama · Llama 3.3 Community · estimated for a 64.0 GB machine at 4K context.

Yes, but it's tight.

Best fit: Llama 3.3 70B Instruct at Q2_K · 26.4 GB · 6–9 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K26.4 GB6–9 tok/sTight / slow
Q3_K_M34.3 GB5–7 tok/sTight / slow
Q4_K_M42.5 GB4–6 tok/sTight / slow
Q5_K_M50.0 GB3–5 tok/sWon't fit
Q6_K57.9 GB3–4 tok/sWon't fit
Q8_075.0 GB2–3 tok/sWon't fit
FP16141.1 GB1–2 tok/sWon't fit

Ran Llama 3.3 70B Instruct on a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB)? Add your real numbers →