Developers

Integrate once. Keep model choice flexible: OpenAI-compatible inference, region-pinned, under one API key.

Your region's endpoint

V1 speaks OpenAI-compatible inference contracts. Point the standard client at your region's /v1 base URL and authenticate with an API key. The example starts on your region — switch it to compare. All credentials shown are placeholders; access opens with the waitlist.

# Region endpoint — OpenAI-compatible /v1 base URL
export CCSIO_AI_BASE_URL=https://api.latam.ccsio.ai/v1
export CCSIO_AI_API_KEY=<your-api-key>            # placeholder, issued with your region

import os
from openai import OpenAI                          # the standard OpenAI client

client = OpenAI(
    base_url=os.environ["CCSIO_AI_BASE_URL"],
    api_key=os.environ["CCSIO_AI_API_KEY"],
)

response = client.chat.completions.create(
    model="qwen-3-8",   # first available model, 320K context
    messages=[{"role": "user", "content": "Hello"}],
)

Model selection

V1 starts with Qwen 3.8 — 320K-token context — in all three regions. Gemma 4 and GLM 5.3 follow on the roadmap, each listing its own availability. Provider-hosted frontier models will always be labeled external, never served as ccsio.ai capacity.

Authentication

You integrate with a per-region API key, issued when your region opens and scoped to that region's endpoint. Until then, every example on this page uses a clearly labeled placeholder — there is no live credential to leak or share.

Governance, by construction

No storage, no sharing, no training on your data. Inference runs on capacity pinned to your region under the 99.999% SLA. Review the full security and sovereignty guarantees below.

Developer FAQ

Is the API OpenAI-compatible?: Yes. V1 is designed around OpenAI-compatible inference contracts, pointed at your region's /v1 base URL with a ccsio.ai API key.
Which endpoint does my region use?: EU: https://api.eu.ccsio.ai/v1, USA: https://api.us.ccsio.ai/v1, LATAM: https://api.latam.ccsio.ai/v1. Your key and your data stay in that region.
What model ids exist in V1?: qwen-3-8 (Qwen 3.8, 320K context) is first in every region. Gemma 4 and GLM 5.3 follow through the waitlist.
Where is my data processed?: Pinned to your chosen region — EU, USA or LATAM — with no storage, sharing or training.
When does access open?: The waitlist is the path: join, pick your region, and your API key arrives when that region opens.

Quickstart — how you get a key

Join the waitlist and pick your region — EU, USA or LATAM.
When your region opens, your API key arrives with your plan.
Point the standard OpenAI client at your region's /v1 endpoint. First model: Qwen 3.8, 320K context.

Get your first API key

The waitlist is the path: account, region, key.