Unverified50% confidenceFactExact time
AI 模型的流式回复通过 Server-Sent Events（SSE）或 WebSocket 实现 token 级推送，OpenAI、Anthropic 等主流 API 均采用这一方式
1
Sources
50%
Confidence
Long-term
Relevance
7/17/2026
First Seen
Sources
Related Claims
UnverifiedMainstream APIs from OpenAI, Anthropic, and others all use the SSE protocol to return token streams, with the frontend consuming data via the EventSource API or the fetch ReadableStream interface.78% similarUnverifiedOne API本质上是一个面向大语言模型的API网关（API Gateway），支持流式响应（Server-Sent Events，SSE）透传和基于Token数量的计量计费76% similarVerifiedMainstream APIs from OpenAI, Anthropic, and others use SSE by default for token-by-token LLM output.76% similarUnverifiedOpenAI's streaming responses use the SSE (Server-Sent Events) protocol and send a data: [DONE] signal when the stream ends.76% similarUnverifiedThe streaming output of Databricks Model Serving is implemented based on the Server-Sent Events (SSE) protocol, and its standard REST API follows a superset of the OpenAI Chat Completions format75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/541167API
curl https://kongchang.com/api/v1/knowledge/claims/541167MCP
get_claim(id=541167)