Skip to content
03 Model garden Serve

gpt oss 120b

Instantly via the EU router or as a dedicated GPU deployment. Data stays in Europe.

curl
$ curl https://hostyourai.com/api/v1/chat/completions \
  -H "Authorization: Bearer hyai-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-oss-120b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Call it now

Drop-in replacement for OpenAI: change only the base URL and API key. The Anthropic format (/v1/messages) is supported too.

Serve Model garden
openai/gpt-oss-120b

text->text · openai · sovereign EU

Runs on EU infrastructure operated by European companies, from your request to the GPU. See the full chain

Welcome to the gpt-oss series, OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.

Router
Router · warm
Dedicated
Dedicated · available
Input (per 1M tokens)
€0.16
Output (per 1M tokens)
€0.63
117BParameters
131KContext window
80GBMinimum VRAM
POST /api/v1/chat/completions 200 OK

Specifications

Parameters117B
Context window131,072 tokens
Context on every source128,000 tokens · The router sends longer requests to a source where they fit.
Minimum VRAM80 GB
ArchitectureGptOssForCausalLM (vLLM)
Licenseapache-2.0
Modalitytext->text
ReleasedAugust 2025
Publisheropenai ↗

Pricing

Shared router · per token

€0.16Input (per 1M tokens)
€0.63Output (per 1M tokens)

Dedicated GPU · per hour

from €2,21 per hourYour own vLLM instance on European cloud (80 GB VRAM), billed per minute.

Shared EU router, you pay only per token. Dedicated GPU deployments are billed per minute, see pricing.

Bench index

Bench is our quality index for every model, 0 to 100. It rests on public measurements (Epoch AI with about 40 benchmarks, LiveBench, LMArena and more) and is refreshed every morning. Where nobody measures, it shows an estimate with a range.

62.8 / 100 · Measured

Measured by Epoch AI across 13 benchmarks, with a range.

Range 60.3 to 65.1 · Rank 126 of 257 at Epoch AI

Sources: Epoch AI ↗ · LiveBench ↗ · LMArena ↗ · Refreshed daily, last on 19-11-2026

LiveBench · 46.1

Reasoning39.2
Coding60.2
Agentic coding16.7
Mathematics68.9
Data analysis38.8
Language48.6
Instruction following50.3
Every benchmark (13)
OTIS Mock AIME 2024-202588.9
GPQA diamond70.5
DTBench60.4
METR Time Horizons56.6
WeirdML48.2
Fiction.LiveBench44.4
Aider polyglot41.8
LMCA26.0
Surface Evolver Bench25.0
Terminal Bench18.7
SimpleBench6.5
APEX-Agents4.4
Mystery Game Puzzles0.0
LiveBench · All tasks
Reasoning
theory of mind48.1
zebra puzzle20.8
spatial80.0
logic with navigation8.0
Coding
code generation70.4
code completion50.0
Agentic coding
javascript20.0
typescript10.0
python20.0
Mathematics
AMPS Hard92.0
integrals with game32.0
math comp92.2
olympiad59.3
Data analysis
consecutive events2.9
tablejoin29.2
tablereformat84.3
Language
connections79.2
plot unscrambling18.6
typos48.0
Instruction following
paraphrase44.8
simplify55.5
story generation56.5
summarize44.4
Questions

Frequently asked questions

Can I run gpt oss 120b in the EU?

Yes. HostYourAI runs gpt oss 120b on GPUs in European datacenters via vLLM. Prompts and outputs never leave the EU and there is no US cloud provider in the chain.

Is hosting gpt oss 120b GDPR-compliant?

Yes. All processing happens inside the EU, a Data Processing Agreement (DPA) is available and the subprocessor list is public. Open-source weights also mean: no training on your data.

How much does gpt oss 120b cost?

Via the shared EU router you pay €0.16 per million input tokens and €0.63 per million output tokens, with no fixed costs. For high volume or isolation you can also run gpt oss 120b on a dedicated GPU instance, billed per minute.

Is the API OpenAI-compatible?

Yes. You use the standard OpenAI SDKs with a custom base URL (https://hostyourai.com/api/v1). The Anthropic Messages API is supported as a drop-in as well.

More models from OpenAI

The platform

One platform, four layers

Start with one layer and add the others when the work asks for it. They share one key, one balance and one Activity view.

Code

Your coding agent on open models in Europe, set up with one command.

Open

Control

Keys, budgets and guardrails per team, and personal data stopped before any model sees it.

Open

Serve

Shared per token, or running for you alone, from European data centres.

Open

EU compute

Private GPUs in Europe, H100 to B300, sized for response time, generation speed and concurrent users.

Open

Try gpt oss 120b for free

Creating an account takes a minute. Test gpt oss 120b straight away in the playground.

  • Jantine
  • Hans
  • Sander
  • Jos
  • Daan
  • Kyra
  • Margriet
  • Stan
The team behind HostYourAI