gpt-oss-120b Hardware Requirements
o4-mini-class reasoning you can self-host. Ships in MXFP4 at ~62GB — beyond any single consumer GPU, but only 5.1B active params make it the staple workload for 128GB Strix Halo mini-PCs and Macs.
VRAM needed (Q4, 8k context)
63.9 GB
No single consumer GPU fits this model — see multi-GPU and Mac options below.
Updated July 2026. Estimates — see methodology below.
VRAM by Quantization
Weights + KV cache at 8k context + 1.2GB system overhead. Q4_K_M is the community default — quality loss is negligible for most use.
| Quantization | Weights | KV cache (8k) | Total VRAM | Cheapest GPU that fits |
|---|---|---|---|---|
| Q4_K_M Recommended — near-lossless for most use, half the size of Q8 | 62.2 GB | 0.6 GB | 63.9 GB | Multi-GPU / Mac territory |
| Q5_K_M Slightly higher quality than Q4 for ~18% more VRAM | 83.4 GB | 0.6 GB | 85.1 GB | Multi-GPU / Mac territory |
| Q8_0 Effectively lossless — use if you have VRAM to spare | 124 GB | 0.6 GB | 126 GB | Multi-GPU / Mac territory |
| FP16 Full precision — only for fine-tuning or maximum fidelity | 234 GB | 0.6 GB | 236 GB | Multi-GPU / Mac territory |
* gpt-oss-120b ships natively quantized (~4.25 bits/weight) — the Q4 row reflects its actual release format.
Longer context costs VRAM
KV cache grows linearly with context: 8k → 0.6 GB · 32k → 2.3 GB · 128k → 9.0 GB. If you plan to feed whole documents or codebases, size your GPU for the context you actually need, not just the weights.
Why gpt-oss-120b is fast but VRAM-hungry
gpt-oss-120b is a Mixture-of-Experts model: all 117B parameters must sit in memory, but each token only activates 5.1B of them. Memory capacity requirements are those of a 117B model, while speed is that of a 5.1B model — which is why MoE models feel so fast when they fit, and why Macs with large unified memory punch above their weight running them.
GPU Compatibility (Q4, 8k context)
Every GPU in our database, scored against gpt-oss-120b. Speed is estimated decode rate — memory-bandwidth-bound, so VRAM and bandwidth matter more than shader count.
| GPU | VRAM | Verdict | Est. speed | Price | |
|---|---|---|---|---|---|
| RTX 3060 | 12 GB | Not enough VRAM | — | ~$238 used | |
| Arc B570 | 10 GB | Not enough VRAM | — | from $225 | |
| RTX 4060 | 8 GB | Not enough VRAM | — | ~$275 used | |
| RX 7600 | 8 GB | Not enough VRAM | — | from $250 | |
| Arc B580 | 12 GB | Not enough VRAM | — | from $250 | |
| RX 6700 XT | 12 GB | Not enough VRAM | — | ~$315 used | |
| Arc A770 | 16 GB | Not enough VRAM | — | from $300 | |
| RTX 4060 Ti | 8 GB | Not enough VRAM | — | ~$338 used | |
| RTX 3070 | 8 GB | Not enough VRAM | — | ~$338 used | |
| RTX 5060 | 8 GB | Not enough VRAM | — | from $325 | |
| RX 7700 XT | 12 GB | Not enough VRAM | — | ~$415 used | |
| RX 6800 XT | 16 GB | Not enough VRAM | — | ~$438 used | |
| RTX 3080 | 10 GB | Not enough VRAM | — | ~$463 used | |
| RX 7800 XT | 16 GB | Not enough VRAM | — | ~$488 used | |
| RTX 4070 | 12 GB | Not enough VRAM | — | ~$500 used | |
| RTX 4070 SUPER | 12 GB | Not enough VRAM | — | ~$563 used | |
| RX 7900 XT | 20 GB | Not enough VRAM | — | ~$588 used | |
| RTX 5060 Ti | 16 GB | Not enough VRAM | — | from $550 | |
| RX 9070 | 16 GB | Not enough VRAM | — | from $575 | |
| RTX 5070 | 12 GB | Not enough VRAM | — | from $600 | |
| RX 9070 XT | 16 GB | Not enough VRAM | — | from $600 | |
| RTX 4070 Ti SUPER | 16 GB | Not enough VRAM | — | ~$750 used | |
| RX 7900 XTX | 24 GB | Not enough VRAM | — | ~$838 used | |
| RTX 4080 SUPER | 16 GB | Not enough VRAM | — | ~$900 used | |
| RTX 5070 Ti | 16 GB | Not enough VRAM | — | from $900 | |
| RTX 3090 | 24 GB | Not enough VRAM | — | ~$1,150 used | |
| RTX 5080 | 16 GB | Not enough VRAM | — | from $1,250 | |
| RTX 4090 | 24 GB | Not enough VRAM | — | ~$2,375 used | |
| RTX 5090 | 32 GB | Not enough VRAM | — | from $2,800 |
Run it in one command
With Ollama installed, this pulls the default quant and starts chatting:
Frequently Asked Questions
How much VRAM do I need to run gpt-oss-120b?+
At the recommended Q4_K_M quantization with 8k context, gpt-oss-120b needs roughly 63.9GB of VRAM (62.2GB weights + KV cache + overhead). Q8 needs about 126GB and full FP16 about 236GB.
Can any single consumer GPU run gpt-oss-120b?+
No single consumer GPU currently has enough VRAM to run gpt-oss-120b fully. Realistic options: a multi-GPU rig (e.g. dual 24GB cards), a Mac with enough unified memory, or a 128GB Ryzen AI Max "Strix Halo" mini-PC — the 2026 favorite for exactly this class of model.
Can I run gpt-oss-120b on an RTX 3060?+
No — 12GB is well below what gpt-oss-120b needs even at Q4 quantization.
Can I run gpt-oss-120b on a Mac?+
Yes, if the Mac has enough unified memory: budget roughly 63.9GB of RAM for the Q4 version (plus what macOS itself uses). Apple Silicon runs GGUF models well via Ollama or LM Studio, and MoE models like this one are particularly Mac-friendly — only 5.1B parameters are active per token, so memory bandwidth goes further.
Can I run gpt-oss-120b on CPU only?+
Sort of. Because only 5.1B of 117B parameters are active per token, CPU inference is more viable than for dense models this size — expect single-digit tokens/sec with fast DDR5. A GPU is still dramatically better.
Is gpt-oss-120b free for commercial use?+
Yes. gpt-oss-120b is released under the Apache 2.0, which permits commercial use.
Related Models
How we calculate these numbers
VRAM = model weights (parameters × bits per weight ÷ 8) + KV cache (architecture-specific bytes per token × context length) + ~1.2GB runtime overhead. Speed estimates assume decode is memory-bandwidth-bound at ~50% utilization (lower for MoE models, which pay routing overhead), matching typical llama.cpp performance on consumer cards; real results vary with runtime, drivers, and settings. Quant sizes reflect GGUF K-quants, which keep some layers at higher precision. Figures are estimates for planning, not guarantees — when in doubt, buy more VRAM than you need today. Prices shown are launch MSRP; mid-2026 street prices often run well above MSRP due to the ongoing memory shortage, and used 24GB cards are holding their value unusually well.