Can I Run Qwen 3.5 on 96 GB system RAM?

Written by Jakub Rusinowski · Last updated February 24, 2026

Yes — comfortably

Yes, comfortably — Qwen 3.5 35B-A3B at Q8_0 needs about 39.9 GB of the 76.8 GB usable on 96 GB system RAM, leaving ~36.9 GB spare and running at ~15.7 tok/s (estimated), with room for about 131,072 tokens of context.

Confidence: medium · Recommended quantization: Q8_0 · Estimated speed: ~15.7 tok/s

96 GB system RAM — what it gives a model

Usable memory for models	76.8 GB
Memory bandwidth	90 GB/s

Qwen 3.5 on 96 GB system RAM: memory by quantization

Quant	Memory needed	Fits 76.8 GB?	Max context	Est. speed	Download
F16	72.7 GB	✓ Yes	16K	~9.5 tok/s	70 GB
Q8_0	39.9 GB	✓ Yes	128K	~15.7 tok/s	37.2 GB
Q6_K	31.4 GB	✓ Yes	128K	~18.9 tok/s	28.7 GB
Q5_K_M	27.5 GB	✓ Yes	128K	~20.9 tok/s	24.8 GB
Q4_K_M	23.8 GB	✓ Yes	128K	~23.2 tok/s	21.1 GB
Q3_K_M	17.6 GB	✓ Yes	128K	~28.4 tok/s	14.9 GB
Q2_K	14.2 GB	✓ Yes	128K	~32.4 tok/s	11.5 GB

Which Qwen 3.5 sizes fit

Variant	Needs at Q4_K_M	Fits?	Est. speed
Qwen 3.5 397B-A17B	244.7 GB	✗ Too large	—
Qwen 3.5 122B-A10B	77.3 GB	✗ Too large	—
Qwen 3.5 35B-A3B	23.8 GB	✓ Fits	~23.2 tok/s
Qwen 3.5 27B	18.8 GB	✓ Fits	~3.9 tok/s
Qwen 3.5 9B	7.4 GB	✓ Fits	~10.9 tok/s
Qwen 3.5 4B	4.1 GB	✓ Fits	~22.3 tok/s
Qwen 3.5 2B	2.7 GB	✓ Fits	~38.9 tok/s
Qwen 3.5 0.8B	1.8 GB	✓ Fits	~74.5 tok/s

What to watch out for

2 larger variants of Qwen 3.5 do not fit and would need CPU offload or different hardware.
This model does not publish its full attention configuration, so the KV-cache share of these figures is inferred from its parameter count rather than computed exactly.
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.
76.8 GB of the 96 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 inferred from the parameter count because this model does not publish its full attention configuration.

FAQ

Can I run Qwen 3.5 on 96 GB system RAM?

Which quantization of Qwen 3.5 should I use on 96 GB system RAM?

Q8_0 — it needs about 39.9 GB of the 76.8 GB available, downloads as roughly 37.2 GB, and runs at an estimated 15.7 tokens/sec with up to 128K of context.

What limits Qwen 3.5 on 96 GB system RAM?

Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.

Can I Run Qwen 3.5 on 96 GB system RAM?

96 GB system RAM — what it gives a model

Qwen 3.5 on 96 GB system RAM: memory by quantization

Which Qwen 3.5 sizes fit

What to watch out for

Recommended setup

How these numbers are calculated

FAQ

Can I run Qwen 3.5 on 96 GB system RAM?

Which quantization of Qwen 3.5 should I use on 96 GB system RAM?

What limits Qwen 3.5 on 96 GB system RAM?

Which runtime should I use?

Other RAM Capacities

Other Models on 96 GB system RAM

Qwen 3.5 on GPUs

What This Model Is Good At

Model & Tools