gpt-oss-120b

ServerlessDedicated
Hugging Face IDopenai/gpt-oss-120bView on Hugging Face ↗

gpt-oss-120b is OpenAI's open-weight model for reasoning, agentic tasks and developer use. It's a 117B-parameter mixture of experts with 5.1B active per token, built to fit on a single 80GB GPU.

Ways to run gpt-oss-120b

Dedicated endpoint

Get a quote →

Your own GPUs, tuned for your traffic. We profile your workload, tune the deployment and put throughput and latency in your contract.

Serverless

API docs →

Pay per token on shared GPUs, with no capacity to plan or reserve.

Inside your network

Security →

Run a dedicated endpoint inside your own network, including air-gapped environments.

Pricing

Dedicated endpoint
Per GPU-hour
We size the GPUs to your traffic.
Serverless
$0.04 / $0.15
Per 1M tokens (input/output)

Use it with the agents we set up for you

Point any open-source agent at your endpoint, or pick from our catalog and we set it up as part of your contract.

Cline CodingOpenCode CodingAider Coding
GPT Researcher ResearchRAGFlow DocumentsNanobrowser Browser

gpt-oss-120b FAQ

How much does gpt-oss-120b cost on Saturn Cloud?

Serverless costs $0.04 per 1M input tokens and $0.15 per 1M output tokens. A dedicated endpoint is priced per GPU-hour for the GPUs it runs on, and we size the GPUs to your traffic.

How do I set the reasoning level in gpt-oss-120b?

Set it to low, medium or high in the system prompt, for example Reasoning: high.

Does gpt-oss-120b support tool calling?

Yes. It supports function calling with schemas, web browsing, Python code execution and Structured Outputs.

Do I need the harmony format for gpt-oss-120b?

Yes. gpt-oss expects OpenAI’s harmony response format, and the chat template applies it for you.

Does gpt-oss-120b fit on one GPU?

Yes. OpenAI built it to fit on a single 80GB GPU, with MXFP4 mixture-of-experts weights. OpenAI also says it can be fine-tuned on a single H100 node.