Access on request
One API. Many models. Under control.
An API compatible with the OpenAI ecosystem to use third-party and open-weight models from a single point. Access is on request.
import os
from openai import OpenAI
client = OpenAI(
base_url="https://criterium.pro/v1",
api_key=os.environ["CRITERIUM_API_KEY"],
)
r = client.chat.completions.create(
model="criterium/router",
messages=[
{"role": "user", "content": "Summarise this invoice."},
],
)
print(r.choices[0].message.content)
Only the marked lines change.
What's included
Five pieces, one API.
- Compatible API
- OpenAI ecosystem clients work by changing the address and the key.
- Models
- Third-party and open-weight. You always know which one answers.
- Routing
- Pick a model, or let a transparent policy choose among several.
- Keys
- Per-organisation keys, shown only once and revocable whenever you want.
- Usage and cost
- Every call metered, with quotas and a budget per organisation.
Request access.
Tell us what you would use it for. We will only write to you about this.