Can I Run Llama 3.3 on Beelink SER9 (Ryzen AI 9, 32 GB)?

Superseded model. Llama 3.3 has been superseded by Llama 4. This page is kept for reference; the newer family is a better starting point. View Llama 4 →

Autor: Jakub Rusinowski · Ostatnia aktualizacja: 8 grudnia 2024

Yes

Yes — Llama 3.3 70B Instruct at Q2_K needs about 26.5 GB of the 32 GB usable on Beelink SER9 (Ryzen AI 9, 32 GB) (~5.5 GB spare), at ~7.7 tok/s (estimated), with room for about 16,384 tokens of context.

Confidence: high · Recommended quantization: Q2_K · Estimated speed: ~7.7 tok/s

Buy This HardwareApple Mac mini M4 (16GB) — 32 GB VRAM · 22 W board power Deploy in the Cloud NowNVIDIA A40 on RunPod — from $0.44/hr · rate checked 2026-08

or compare on Vast.ai

As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.

Beelink SER9 (Ryzen AI 9, 32 GB) — what it gives a model

Usable memory for models	32 GB
Memory bandwidth	256 GB/s
Form factor	Mini PC
Operating system	Windows or Linux
Memory upgradeable	Yes
Price	$859 (lib/data/ai-stations.ts (street price), checked 2026-07-06)

Llama 3.3 on Beelink SER9 (Ryzen AI 9, 32 GB): memory by quantization

Quant	Memory needed	Fits 32 GB?	Max context	Est. speed	Download
F16	143.5 GB	✗ No	—	—	140 GB
Q8_0	77.9 GB	✗ No	—	—	74.4 GB
Q6_K	60.9 GB	✗ No	—	—	57.4 GB
Q5_K_M	53.1 GB	✗ No	—	—	49.6 GB
Q4_K_M	45.7 GB	✗ No	—	—	42.3 GB
Q3_K_M	33.3 GB	✗ No	—	—	29.8 GB
Q2_K	26.5 GB	✓ Yes	16K	~7.7 tok/s	23 GB

What to watch out for

Q2_K is the only quantization that fits, and it is a heavily degraded one — expect noticeably worse output than the same model at Q4_K_M. A smaller model at Q4 is usually the better trade.
1 larger variant of Llama 3.3 does not fit and would need CPU offload or different hardware.

Beelink SER9 32 GB limitations

Shares system memory with the iGPU, so the usable model budget is well below the nominal 32 GB.
Memory bandwidth, not capacity, is the limit here — expect single-digit tokens/sec on large models.

Recommended setup

Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving

How these numbers are calculated

Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
32 GB unified memory at 256 GB/s, shared between CPU and GPU.
Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
KV cache is computed from this model's published attention configuration.

FAQ