GETTING STARTED
CLSSAI API quickstart
Use one OpenAI-compatible endpoint for open and frontier models. Keep your existing SDK and change the base URL.
This first documentation surface mirrors the intended product contract. Protocol edge cases and generated API reference remain in the workbench.
1. Create an API Key
Sign in, open your workspace, and create a scoped inference key. Production keys will be shown once and only their keyed hash is stored.
Open API Keys2. Send a request
from openai import OpenAIclient = OpenAI( base_url="https://api.clssai.com/v1", api_key=os.environ["CLSSAI_API_KEY"],)response = client.chat.completions.create( model="deepseek/deepseek-v4-flash-0731", messages=[{"role": "user", "content": "Hello"}])3. Inspect the result
- Actual routed model and Provider
- Request ID and policy version
- Token use and settled cost
- Retry attempts before streaming
Next: explore models
Filter by capability, modality, official price, context, and data policy.