GPT-oss 120B — VRAM Requirements
Autor: Jakub Rusinowski · Ostatnia aktualizacja: 21 lipca 2026
How much GPU VRAM you need to run GPT-oss 120B GPT-oss 120B by OpenAI locally, a 120B-parameter model. Figures are quantized weights + KV cache + framework overhead from a memory-bandwidth roofline model.
GPT-oss 120B needs about 73 GB VRAM at Q4_K_M.
VRAM by Quantization
| Quant | Bits/weight | Weights | Total VRAM |
|---|---|---|---|
| Q2_K | 2.63 | 39.4 GB | 40.2 GB |
| Q3_K_M | 3.41 | 51.2 GB | 52.0 GB |
| Q4_K_M | 4.83 | 72.5 GB | 73.3 GB |
| Q5_K_M | 5.67 | 85.0 GB | 85.8 GB |
| Q6_K | 6.56 | 98.4 GB | 99.2 GB |
| Q8_0 | 8.50 | 127.5 GB | 128.3 GB |
| F16 | 16.00 | 240.0 GB | 240.8 GB |
Switch quantization in the interactive calculator, or see the full GPT-oss 120B model page.
Add this badge to your model card
Model creators: paste this into your Hugging Face model card README to link readers straight to this VRAM breakdown.
[](https://llmconfigurator.com/en/vram-calculator/gpt-oss-120b-full)Estimates only — actual VRAM varies with context length, batch size, runtime and KV-cache settings.