Chat Completions

The Chat Completions endpoint is OpenAI-compatible. Use it with OpenAI models and any OpenAI-protocol model in the catalog.

POST https://api.smartapi.cc/api/openapi/v1/chat/completions

Authentication#

Authorization: Bearer sk-your-key-here
Content-Type: application/json

Request body#

SmartAPI proxies the request body to the upstream provider, so all standard OpenAI Chat Completions parameters are supported. The most common fields:

Field	Type	Required	Description
`model`	string	Yes	Model identifier, e.g. `gpt-4.1`.
`messages`	array	Yes	Conversation messages, each with `role` and `content`.
`stream`	boolean	No	Stream the response as server-sent events. Defaults to `false`.
`temperature`	number	No	Sampling temperature.
`max_tokens`	integer	No	Maximum tokens to generate.

Only model is validated by SmartAPI itself; all other parameters are forwarded to the upstream provider unchanged.

Whether to enable thinking/reasoning is controlled by request parameters. Models that support thinking carry a thinking tag on the Models page. Thinking is off by default when none of the parameters below are sent.

Field	Type	Description
`reasoning_effort`	string	Reasoning level: `none` / `minimal` / `low` / `medium` / `high` / `xhigh`. `none` disables it; any other value enables it.
`thinking`	object	Extended thinking config (aligned with Anthropic / Bedrock Claude). When present alongside `reasoning_effort`, `thinking.type` takes precedence.

thinking object fields:

Field	Type	Description
`type`	string	`enabled` / `disabled`.
`budget_tokens`	number	Thinking token budget; ensure `max_tokens > budget_tokens`.
`display`	string	e.g. `omitted`; forwarded when the upstream supports it.

{
  "model": "gpt-5.5",
  "messages": [{ "role": "user", "content": "Hello" }],
  "reasoning_effort": "medium"
}

Basic example#

curl https://api.smartapi.cc/api/openapi/v1/chat/completions \
  -H "Authorization: Bearer $SMARTAPI_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4.1",
    "messages": [
      { "role": "system", "content": "You are a helpful assistant." },
      { "role": "user", "content": "What is SmartAPI?" }
    ]
  }'

Response#

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "created": 1730000000,
  "model": "gpt-4.1",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "SmartAPI is..." },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 24,
    "completion_tokens": 88,
    "total_tokens": 112
  }
}

Streaming#

Set stream to true to receive the response incrementally as text/event-stream server-sent events. SmartAPI automatically requests usage data in the final chunk.

from openai import OpenAI
 
client = OpenAI(
    api_key="sk-your-key-here",
    base_url="https://api.smartapi.cc/api/openapi/v1",
)
 
stream = client.chat.completions.create(
    model="gpt-4.1",
    messages=[{"role": "user", "content": "Write a haiku about APIs."}],
    stream=True,
)
for chunk in stream:
    delta = chunk.choices[0].delta.content
    if delta:
        print(delta, end="")

Each SSE chunk looks like:

data: {"id":"chatcmpl-...","choices":[{"delta":{"content":"Hello"}}]}

data: [DONE]

The stream terminates with a data: [DONE] line.

Usage and billing#

The usage object reports prompt_tokens, completion_tokens, and total_tokens. When the provider supports prompt caching, usage may also include prompt_tokens_details.cached_tokens. Token usage drives billing.