Chat Completions

OpenAI-compatible POST /v1/chat/completions Live. Supports model, messages, stream, temperature (0–2), max_tokens, top_p, stop.

curl https://ciptamodel.com/v1/chat/completions \
  -H "Authorization: Bearer sk-cm-live-..." \
  -H "Content-Type: application/json" \
  -d '{"model": "deepseek-v4.1-flash", "temperature": 0.7,
       "messages": [{"role": "system", "content": "You are concise."},
                    {"role": "user", "content": "Explain rate limiting."}]}'

Non-streaming returns a chat.completion object with real provider usage (prompt/completion/total tokens). Token counts come from the provider — never fabricated. A cost estimate appears when the model registry carries pricing.