gpt-oss-120b
OpenAI’s larger open-weight model: o4-mini-class reasoning you could also run yourself.
About gpt-oss-120b
gpt-oss-120b was OpenAI’s first open-weight language model since GPT-2, released in August 2025 under Apache 2.0. OpenAI says it comes close to o4-mini on core reasoning tests and beats it on competition maths and health questions, and it scores 80.1% on GPQA Diamond. It was built to fit on a single 80GB GPU.
It reasons at low, medium or high effort and shows its full chain of thought. Its small active size makes it inexpensive to run, which makes it a sensible choice for maths, logic and structured tasks at volume.
Where it’s strong
Maths, science and logic problems.
Structured, step-by-step tasks at volume.
Agent steps that call functions.
Where to be careful
Its knowledge stops at June 2024, the oldest of the reasoning models here.
Text only.
Price
For scale: 1,000 requests of about 3,000 tokens in and 1,000 out cost $0.37 — a long prompt and a page of answer each. Every rate is published and stays the same from one request to the next. See every model.
Specs
gpt-oss-120bCall it
Use the OpenAI SDK you already have — point it at Roar AI and name the model. Read the docs.
from openai import OpenAI
client = OpenAI(base_url="https://api.roar-ai.com/v1", api_key="roar_live_…")
reply = client.chat.completions.create(
model="gpt-oss-120b",
messages=[{"role": "user", "content": "Summarise this contract in plain English."}],
)Compare gpt-oss-120b
Questions
How much does gpt-oss-120b cost?
Input is $0.0481 per 1M tokens; output is $0.221 per 1M tokens. For scale, 1,000 requests of about 3,000 tokens in and 1,000 out cost $0.37. The rate is published and does not change from one request to the next.
Can I use gpt-oss-120b alongside other models?
Yes. Every model on Roar AI is on the same API key and the same monthly invoice, so switching from gpt-oss-120b to another model is a change to one word in your code.
Is gpt-oss-120b open source?
Its weights are open under the Apache 2.0 licence, so you could run it yourself. Through Roar AI you call it like any other model, with no servers to manage.