流式响应

通过 Server-Sent Events 接收 Chat Completions 的增量输出。

设置 stream: true 后，接口以 Server-Sent Events（SSE）返回多个 chat.completion.chunk 事件。每个事件中的增量文本位于 choices[0].delta.content。

curl --no-buffer https://api.tikway.ai/v1/chat/completions \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5.6-terra",
    "stream": true,
    "messages": [
      {
        "role": "user",
        "content": "用三句话描绘雨后的城市。"
      }
    ]
  }'

​

事件示例

data: {"id":"chatcmpl_01J...","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant","content":"雨停"},"finish_reason":null}]}

data: {"id":"chatcmpl_01J...","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"之后，霓虹"},"finish_reason":null}]}

data: {"id":"chatcmpl_01J...","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: [DONE]

客户端应持续拼接非空的 delta.content，直到收到 [DONE]。不要假设每个事件都包含文本：首个事件可能只包含 role，结束事件则只包含 finish_reason。

​

返回用量

如需在流结束时获取用量，可在请求中加入：

{
  "stream_options": {
    "include_usage": true
  }
}