# gpt-oss-120b GPU requirements

gpt-oss-120b has 117B parameters. To serve it, plan for about 280 GB of GPU memory in 16-bit, 140 GB in 8-bit or 70 GB in 4-bit, including about 20% for the KV cache and activations.

It is a mixture of experts: 128 experts, 4 of them active for each token. All of them stay in GPU memory, so plan memory for all 117B parameters.

[Deploy it on Arrakis](/console/#/models?model=openai/gpt-oss-120b).

## What it takes to serve it

| Precision | Memory and the smallest setup |
|---|---|
| 16-bit | 280 GB: 2 × [H200](/gpus/h200.html.md), from $8.06 per hour for all 2 |
| 8-bit | 140 GB: one [H200](/gpus/h200.html.md), from $4.03 per GPU-hour |
| 4-bit | 70 GB: one [H100](/gpus/h100.html.md) or [A100](/gpus/a100.html.md), H100 from $1.65 per GPU-hour |

## What it takes to fine-tune it

| Method | Memory and the smallest setup |
|---|---|
| QLoRA | 82 GB: one [RTX PRO 6000](/gpus/rtx-pro-6000.html.md), from $1.90 per GPU-hour |
| LoRA | 280 GB: 2 × [H200](/gpus/h200.html.md), from $8.06 per hour for all 2 |
| Full fine-tuning | 1,869 GB: 8 × [B300](/gpus/b300.html.md), from $36.80 per hour for all 8 |

## About the model

| | |
|---|---|
| Parameters | `117B` |
| Experts | `128` |
| Experts for each token | `4` |
| Context window (tokens) | `131,072` |
| Published in | `FP8` |
| License | `apache-2.0` |
| Released | `2025-08-04` |
| Weights | [openai/gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) |

## Serve it yourself

```
vllm serve openai/gpt-oss-120b --tensor-parallel-size 2
```

## Questions

**How much GPU memory does gpt-oss-120b need?** To serve it, about 280 GB in 16-bit, 140 GB in 8-bit, 70 GB in 4-bit, including about 20% for the KV cache and activations. To fine-tune it, about 82 GB with QLoRA, 280 GB with LoRA and 1,869 GB for full fine-tuning with Adam, before activations.

**Can I run gpt-oss-120b on one GPU?** Yes, quantized. In 8-bit it needs about 140 GB, which fits on one H200. In 16-bit it needs 2 H200 GPUs.

**What does it cost to run gpt-oss-120b?** In 16-bit it needs 2 H200 GPUs; in 4-bit, one H100 or A100. Each setup on this page shows the lowest price per GPU-hour on Arrakis right now, and you pay by the hour from a prepaid balance.

**How do I deploy gpt-oss-120b on Arrakis?** Open Models in the Arrakis console, paste openai/gpt-oss-120b, choose vLLM, SGLang or llama.cpp, and rent a GPU it fits on. You get an OpenAI-compatible endpoint with its own API key, and root SSH to the machine.


More models: [/models](/models.html.md).