Can I Run GPT-OSS on 192 GB system RAM?

作者： Jakub Rusinowski · 最后更新： 2026年8月15日

Yes

Yes — GPT-oss 120B at Q8_0 needs about 125.5 GB of the 153.6 GB usable on 192 GB system RAM (~28.1 GB spare), at ~11.4 tok/s (estimated), with room for about 131,072 tokens of context.

Confidence: high · Recommended quantization: Q8_0 · Estimated speed: ~11.4 tok/s

Buy This HardwareApple Mac Studio M4 Max — 128 GB VRAM · 35 W board power Deploy in the Cloud NowRunPod

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.

192 GB system RAM — what it gives a model

Usable memory for models	153.6 GB
Memory bandwidth	90 GB/s

GPT-OSS on 192 GB system RAM: memory by quantization

Quant	Memory needed	Fits 153.6 GB?	Max context	Est. speed	Download
F16	235 GB	✗ No	—	—	233.6 GB
Q8_0	125.5 GB	✓ Yes	128K	~11.4 tok/s	124.1 GB
Q6_K	97.2 GB	✓ Yes	128K	~14.5 tok/s	95.8 GB
Q5_K_M	84.2 GB	✓ Yes	128K	~16.5 tok/s	82.8 GB
Q4_K_M	71.9 GB	✓ Yes	128K	~19 tok/s	70.5 GB
Q3_K_M	51.2 GB	✓ Yes	128K	~25.6 tok/s	49.8 GB
Q2_K	39.8 GB	✓ Yes	128K	~31.6 tok/s	38.4 GB

Which GPT-OSS sizes fit

Variant	Needs at Q4_K_M	Fits?	Est. speed
GPT-oss 120B	71.9 GB	✓ Fits	~19 tok/s
GPT-OSS 20B	13.8 GB	✓ Fits	~26.6 tok/s

What to watch out for

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.
153.6 GB of the 192 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