Can I Run GPT-OSS on 192 GB system RAM?
Written by Jakub Rusinowski · Last updated August 15, 2026
Technically yes, but not recommended
It loads, but it is not worth running — GPT-oss 120B at Q8_0 fits in 192 GB system RAM's 153.6 GB, yet the memory bandwidth limits it to ~0.5 tok/s (estimated), well below usable interactive speed.
Confidence: high · Recommended quantization: Q8_0 · Estimated speed: ~0.5 tok/s
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 | 241.4 GB | ✗ No | — | — | 240 GB |
| Q8_0 | 128.9 GB | ✓ Yes | 64K | ~0.5 tok/s | 127.5 GB |
| Q6_K | 99.8 GB | ✓ Yes | 64K | ~0.7 tok/s | 98.4 GB |
| Q5_K_M | 86.5 GB | ✓ Yes | 64K | ~0.8 tok/s | 85.1 GB |
| Q4_K_M | 73.9 GB | ✓ Yes | 64K | ~0.9 tok/s | 72.5 GB |
| Q3_K_M | 52.6 GB | ✓ Yes | 64K | ~1.3 tok/s | 51.2 GB |
| Q2_K | 40.9 GB | ✓ Yes | 64K | ~1.7 tok/s | 39.4 GB |
Which GPT-OSS sizes fit
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|
| GPT-oss 120B | 73.9 GB | ✓ Fits | ~0.9 tok/s |
| GPT-OSS 20B | 13.3 GB | ✓ Fits | ~5.4 tok/s |
What to watch out for
- At ~0.5 tok/s this loads but is too slow for interactive use — expect roughly 120 seconds per 60 tokens.
- 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
Can I run GPT-OSS on 192 GB system RAM?
It loads, but it is not worth running — GPT-oss 120B at Q8_0 fits in 192 GB system RAM's 153.6 GB, yet the memory bandwidth limits it to ~0.5 tok/s (estimated), well below usable interactive speed.
Which quantization of GPT-OSS should I use on 192 GB system RAM?
Q8_0 — it needs about 128.9 GB of the 153.6 GB available, downloads as roughly 127.5 GB, and runs at an estimated 0.5 tokens/sec with up to 64K of context.
What limits GPT-OSS on 192 GB system RAM?
Memory bandwidth. The model fits, but at 89.6 GB/s it can only be read fast enough for roughly 0.5 tokens/sec.
Which runtime should I use?
llama.cpp (CPU build) or Ollama — both run without a GPU
Other RAM Capacities
Other Models on 192 GB system RAM
GPT-OSS on GPUs
What This Model Is Good At
Model & Tools
← Can I Run It? | GPT-OSS model page | Check your hardware