Can I Run Qwen3-Coder on 64 GB system RAM?

Autor: Jakub Rusinowski · Ostatnia aktualizacja: 29 września 2026

Yes, but it is tight

Yes, but it is tight — Qwen3-Coder-Next (80B-A3B MoE) at Q4_K_M needs about 49.4 GB of the 51.2 GB usable on 64 GB system RAM, leaving only ~1.8 GB before the runtime starts swapping. Expect ~31.4 tok/s (estimated), with room for about 65,536 tokens of context.

Confidence: medium · Recommended quantization: Q4_K_M · Estimated speed: ~31.4 tok/s

Get a personalized upgrade path →

Kup ten sprzęt NVIDIA DGX Spark (128GB) — 128 GB VRAM · 150 W board power Uruchom w chmurze NVIDIA A100 80GB w RunPod — od $1.39/godz. · stawka sprawdzona 2026-07

lub porównaj w Vast.ai od $0.77/godz. (typical low · varies)

Jako partner Amazon zarabiamy na kwalifikujących się zakupach. Linki do chmurowych GPU są linkami polecającymi — możemy otrzymać prowizję bez dodatkowych kosztów dla Ciebie.

64 GB system RAM — what it gives a model

Usable memory for models	51.2 GB
Memory bandwidth	90 GB/s

Qwen3-Coder on 64 GB system RAM: memory by quantization

Quant	Memory needed	Fits 51.2 GB?	Max context	Est. speed	Download
F16	161.1 GB	✗ No	—	—	160 GB
Q8_0	86.1 GB	✗ No	—	—	85 GB
Q6_K	66.7 GB	✗ No	—	—	65.6 GB
Q5_K_M	57.8 GB	✗ No	—	—	56.7 GB
Q4_K_M	49.4 GB	✓ Yes	64K	~31.4 tok/s	48.3 GB
Q3_K_M	35.2 GB	✓ Yes	256K	~41.8 tok/s	34.1 GB
Q2_K	27.4 GB	✓ Yes	256K	~51 tok/s	26.3 GB

Which Qwen3-Coder sizes fit

Variant	Needs at Q4_K_M	Fits?	Est. speed
Qwen3-Coder 480B-A35B (MoE)	291.6 GB	✗ Too large	—
Qwen3-Coder-Next (80B-A3B MoE)	49.4 GB	✓ Fits	~31.4 tok/s
Qwen3-Coder 30B-A3B (MoE)	20 GB	✓ Fits	~26.4 tok/s

What to watch out for

1 larger variant of Qwen3-Coder does not fit and would need CPU offload or different hardware.
These figures assume CPU-only inference. Any discrete GPU, even an 8 GB one, will be several times faster for models that fit in its VRAM.

Recommended setup

llama.cpp (CPU build) or Ollama — both run without a GPU

How these numbers are calculated

Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
51.2 GB of the 64 GB is treated as usable for model weights (80% — the rest is the OS and running applications).
DDR5-5600 dual channel at 89.6 GB/s peak. CPU decode is assumed to sustain 35% of that peak, because CPU inference is not purely bandwidth-bound — it also spends real time in compute and thread synchronisation. This figure is an assumption, not a fitted constant: no CPU measurement is in the calibration set.
CPU-only inference: no GPU is assumed. A GPU of any size will beat these figures substantially.
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