razorBridge AI Gateway
AI models by the token. You choose where they run.
One OpenAI-compatible API, with a price per million tokens and a residency label on every model. EU-only models run in Paris on an EU-headquartered provider and never fall back outside the EU.
From €0,15 per 1M input tokens · Mistral Small 3.2, EU-only
Illustrative photograph
How it works
Change the base URL. Keep your code.
Models are grouped by where they run. EU-only models run in Paris with an EU-headquartered provider and fail closed, with no fallback outside the EU. Every other model says so and names its provider.
- Input and output price per million tokens on every model.
- Self-serve keys draw on the same prepaid balance as your AI Blades.
- At cost up to 5.000.000 tokens a month, then +5%.
# python -m pip install openai
import os
from openai import OpenAI
client = OpenAI(
base_url="https://razorbridge.eu/api/v1/compute/gate",
api_key=os.environ["RAZORBRIDGE_KEY"],
max_retries=0,
)
response = client.chat.completions.create(
model="razorbridge-eu/deepseek-v4-flash-0731",
messages=[{"role": "user", "content": "Hello!"}],
max_tokens=128,
)
print(response.choices[0].message.content)curl --fail-with-body https://razorbridge.eu/api/v1/compute/gate/chat/completions \
-H "Authorization: Bearer $RAZORBRIDGE_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"razorbridge-eu/deepseek-v4-flash-0731","messages":[{"role":"user","content":"Hello!"}],"max_tokens":128}'Drop-in replacement for any OpenAI-compatible client.
01
Create a keyCreate an API key in the web app or through the REST API.
02
Point your SDKUse the standard OpenAI SDK with our endpoint.
03
Pay per tokenCredits deducted per token, automatically.
Models and pricing · Quickstarts · EU API checklist
Models
Every model, with where it runs.
Per 1M tokens, excl. VAT, from the public model catalogue.
| Model | Processed by | Price / 1M tokens | Try |
|---|---|---|---|
|
EU-only
Paris, France · EU-headquartered provider · no fallback outside the EU |
|||
| DeepSeek V4 Flash | EU model provider (France) | In €0,40 · Out €0,80 | Try |
| Gemma 4 26B | EU model provider (France) | In €0,25 · Out €0,50 | Try |
| GLM 5.2 | EU model provider (France) | In €1,80 · Out €5,50 | Try |
| GPT-OSS 120B | EU model provider (France) | In €0,15 · Out €0,60 | Try |
| Llama 3.3 70B | EU model provider (France) | In €0,90 · Out €0,90 | Try |
| Mistral Medium 3.5 | EU model provider (France) | In €1,50 · Out €7,50 | Try |
| Mistral Small 3.2 | EU model provider (France) | In €0,15 · Out €0,35 | Try |
| Qwen3 235B | EU model provider (France) | In €0,75 · Out €2,25 | Try |
| Qwen3.5 397B | EU model provider (France) | In €0,60 · Out €3,60 | Try |
| Qwen3.6 35B | EU model provider (France) | In €0,25 · Out €1,50 | Try |
| Qwen3.8 27B | EU model provider (France) | In €0,60 · Out €3,30 | Try |
|
May be processed outside the EU
Models marked “May be processed outside the EU” send your prompts and outputs through our US routing provider, listed in our sub-processors, to third-party model providers, some headquartered and processing outside the EU. Do not use them for data that must stay in the EU. Models outside the EU are routed only to providers that don't retain or train on your data. See who processes razorBridge AI Gateway requests |
|||
| Qwen 2.5 7B Instruct | US routing provider (United States) | In €0,10 · Out €0,19 | Try |
| GPT-4o | openai | In €2,30 · Out €9,20 | Try |
With AI Blades
Train on a GPU, serve through the API.
One razorBridge account covers both.
SSH into an AI Blade and fine-tune or train on your own data.
Create an API key and query any model via the OpenAI SDK. Credit-billed per token.
Self-serve keys draw on the same prepaid balance as your AI Blades.
Questions
Straight answers.
Something else? Talk to us.
Which AI models stay in the EU?
The ones labelled EU-only. They run in Paris with an EU-headquartered provider and fail closed rather than fall back outside the EU.
How does billing work?
Self-serve accounts use a prepaid balance. Each GPU bills at least 60 minutes, then per second, and AI models bill per million tokens. Universities and contracts pay by invoice. You see the price before every start.
Get started
Change the base URL. Keep your code.
Create a key and send your first request.