What GPU do I need to run openai/gpt-oss-120b?
A 117B (MoE) reasoning model whose MXFP4 weights fit a single 80GB GPU. 116.8B parameters, published in MXFP4. View on Hugging Face
What gpt-oss-120b is
gpt-oss-120b is a 117B-parameter mixture-of-experts language model published by OpenAI on Hugging Face, one of two open-weight gpt-oss models built for reasoning, agentic tasks and developer use. Its MoE weights are post-trained in MXFP4 precision, which the model card says is what lets a 117B-parameter model run on a single 80GB GPU (NVIDIA H100 or AMD MI300X). It supports configurable reasoning effort (low, medium, high), function calling, web browsing and Python code execution, and is released under Apache 2.0.
License note: fully permissive, no gating and no usage restrictions. Facts in this section are sourced from gpt-oss-120b's Hugging Face model card, not benchmarked by Aquanode.
What it's used for
- Reasoning and chain-of-thought tasks
- Agentic tool use and function calling
- Web browsing and code-execution agents
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision, plus a fixed overhead for KV-cache, activations, and fragmentation. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| MXFP4 (native) | 60.8 GB | 80.0 GB | A100 | 1 | $1.31/hr |
| cheaper alt. | RTX 5060 Ti | 5 | $0.550/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
OpenAI's own model card: MXFP4-quantized MoE weights let gpt-oss-120b run on a single 80GB GPU (NVIDIA H100 or AMD MI300X).
Cheapest way to run gpt-oss-120b at its published (MXFP4) precision: 1× A100, at $1.31/hr per GPU ($1.31/hr total).
How to run gpt-oss-120b
Run gpt-oss-120b with vLLM
From openai/gpt-oss-120b's own model card (requires vllm>=0.10.1+gptoss).
vllm serve openai/gpt-oss-120bSource: https://huggingface.co/openai/gpt-oss-120b/raw/main/README.md
Run gpt-oss-120b with Ollama
From the model card's own Ollama instructions; verified against Ollama's own library listing.
ollama run gpt-oss:120bRun gpt-oss-120b with GGUF quantizations
Prebuilt GGUF weights published at unsloth/gpt-oss-120b-GGUF. Run with llama.cpp's llama-server or load the repo directly in LM Studio.
llama-server -hf unsloth/gpt-oss-120b-GGUFDeploy gpt-oss-120b on Aquanode
Aquanode has no one-click deploy template for gpt-oss-120b; you install the inference engine yourself with the commands below. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch a bare GPU pod sized to the requirement above (1× A100 or larger).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch a GPU podWeight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More gpt-oss models
- gpt-oss-20b-BF16 (20.9B, BF16)
- Huihui-gpt-oss-20b-BF16-abliterated (20.9B, BF16)
- gpt-oss-20b (20.9B, MXFP4)
- gpt-oss-120b-Eagle3-short-context (787M, BF16)
- EAGLE3-gpt-oss-20b (350M, BF16)
Related reading: A100 pricing and specs, and The best GPUs for AI, ranked.
