List your models
curl "$CCSIO_BASE_URL/models" \
-H "Authorization: Bearer $CCSIO_API_KEY"JSON
{
"object": "list",
"data": [
{ "id": "gemma-4-31b", "object": "model", "created": 1791350400, "owned_by": "ccsio", "capabilities": ["chat-completions"] },
{ "id": "kokoro-82m", "object": "model", "created": 1791350400, "owned_by": "ccsio", "capabilities": ["audio-speech"] },
{ "id": "qwen3-embedding-0.6b", "object": "model", "created": 1791350400, "owned_by": "ccsio", "capabilities": ["embeddings"] },
{ "id": "qwen3.8-flash-next", "object": "model", "created": 1791350400, "owned_by": "ccsio", "capabilities": ["chat-completions"] },
{ "id": "whisper-large-v3", "object": "model", "created": 1791350400, "owned_by": "ccsio", "capabilities": ["audio-transcriptions"] }
]
}capabilities.Capabilities and endpoints
| Capability | Endpoint | Guide |
|---|---|---|
chat-completions | POST /v1/chat/completions | Chat completions |
responses | POST /v1/responses | Responses |
embeddings | POST /v1/embeddings | Embeddings |
audio-speech | POST /v1/audio/speech | Audio |
audio-transcriptions | POST /v1/audio/transcriptions | Audio |
image-generations | POST /v1/images/generations | Images |
decisions | POST /v1/systemone | Decisions |
A model called on an endpoint it does not serve is refused with 404 model-not-found. To pick a model for a task, filter the list by capability:
import os
from openai import OpenAI
client = OpenAI(
base_url=os.environ["CCSIO_BASE_URL"],
api_key=os.environ["CCSIO_API_KEY"],
)
def models_for(capability: str) -> list[str]:
"""The ids of the models your organization can call for one capability."""
return [m.id for m in client.models.list() if capability in m.model_extra.get("capabilities", [])]
print(models_for("chat-completions"))
print(models_for("embeddings"))Included and additional models
- Included models are served by CCSolutions.io and come with your organization's plan. They are not charged per request.
- Additional models are optional models CCSolutions.io offers on top, paid from your prepaid credits at a price per million tokens. See them with their prices in the console under Models → Additional models, and your balance under Billing → Credits.
Models of your own providers
An administrator can connect an OpenAI-compatible provider your organization already pays for, with its own key, under Models → Your providers. Its models are called like any other, with the id the console shows (byo:<organization>:<name>), and the provider bills them.
Models per team
Administrators can restrict a model to some teams. A key that acts for a team (see Acting for a team) lists and calls only the models that team may use.