Inference.
On your terms.

Open-weight models on UK & EU infrastructure.

Three ways to run.

Compare services

Access open-weight models on a shared fleet. Pay for the tokens you use.

Explore Token API

Choose your region.

UK or EU processing. Your choice of retention and isolation.

No training on your data.

Prompts and outputs stay out of model training.

Your retention policy.

Storage and caching are your choice.

Open-weight models.

A familiar API and a choice of models.

Your data & inference

Keep your existing tools.

An OpenAI-compatible API for your application.

Explore the platform
from openai import OpenAI

client = OpenAI(
    base_url="YOUR_INFERFOUNDRY_ENDPOINT",
    api_key="YOUR_API_KEY",
)

response = client.chat.completions.create(
    model="YOUR_MODEL_ID",
    messages=[{"role": "user", "content": "Hello"}],
)

Endpoint, key and model ID are placeholders.