gpt-oss-120b VRAM requirements
OpenAI · 117B params (5.1B active) · Apache 2.0. Estimates, not vendor numbers.
gpt-oss-120b (117B parameters, 5.1B active per token) needs about 72 GB of VRAM or unified memory at Q4, 85 GB at Q5 and 126 GB at Q8 (weights plus about 1.5 GB headroom, our estimate). The smallest common setup that runs it is Mac 128GB (M4 Max).
| Quant | Download | Memory needed |
|---|---|---|
| Q4 | 70 GB | 72 GB |
| Q5 | 83 GB | 85 GB |
| Q8 | 124 GB | 126 GB |
Will it run on my machine?
| Setup | Runs? | Speed est. |
|---|---|---|
| 8GB GPU (RTX 4060) | No | n/a |
| 12GB GPU (RTX 3060 / 4070) | No | n/a |
| 16GB GPU (RTX 4060 Ti 16GB) | No | n/a |
| 24GB GPU (RTX 4090) | No | n/a |
| 32GB GPU (RTX 5090) | No | n/a |
| Mac 16GB (M4) | No | n/a |
| Mac 48GB (M4 Pro) | No | n/a |
| Mac 128GB (M4 Max) | Yes, Q5 | ~90 tok/s |
| PC, 32GB RAM, no GPU | No | n/a |
Best quant that fits each setup, with estimated generation speed from memory bandwidth. Real speed depends on the runtime, context length and settings.
Mixture of experts: 117B total, 5.1B active per token, so it runs fast for its size.
On the best local model board (LMArena, CC BY 4.0) it ranks #4 among the open models in our finder.
FAQ
How much VRAM does gpt-oss-120b need?
About 72 GB at Q4, 85 GB at Q5 and 126 GB at Q8, including about 1.5 GB headroom. These are estimates from the parameter count on the official model card.
Can gpt-oss-120b run on 8GB of VRAM?
Not fully on the GPU. At Q4 it needs about 72 GB; partial CPU offload works but is much slower.
Can gpt-oss-120b run on a Mac?
Yes. Mac 128GB (M4 Max) can run it.
Which quant of gpt-oss-120b should I use?
Use the highest quant that fits with room to spare: Q8 is near full quality, Q5 is a good balance, and Q4 is the smallest that usually keeps quality acceptable.
Why is gpt-oss-120b fast for its size?
It is a mixture of experts: 117B parameters in total but only 5.1B are used per token, so it reads less memory per token. You still need memory for all 117B.
Similar-size models
Gemma 4 31B (30.7B) · DeepSeek R1 Distill Qwen 32B (32.8B) · Qwen3.6 35B A3B (35B) · Llama 3.3 70B Instruct (70B) · Mistral Small 4 (119B) · Qwen3.8 Flash Next (180B)