Unverified50% confidenceFactExact time
早期与大模型交互的方式本质上是 RESTful API 调用，开发者通过 HTTP POST 请求将提示词发送到模型服务端点，服务端返回 JSON 格式的生成结果
1
Sources
50%
Confidence
Long-term
Relevance
8/23/2026
First Seen
Sources
Related Entities
Related Claims
Unverified大语言模型的API接口（如OpenAI的Chat Completions API、Anthropic的Messages API）通常采用HTTP REST协议，返回JSON格式数据，并支持Server-Sent Events流式输出76% similarVerifiedRESTful API uses URLs to represent resources, HTTP methods to represent operations (GET for query, POST for creation, PUT for update, DELETE for deletion), and HTTP status codes to represent results.69% similarUnverifiedOpenAPI Specification（Swagger）本质上使用 JSON Schema 的子集来描述 HTTP 接口的请求和响应结构69% similarUnverifiedUnder the tool-calling paradigm, the LLM outputs a structured function-call intent (in JSON format), which is actually executed by an external system that returns the result to the model.69% similarUnverifiedPrompt API 配合 JSON Schema 结构化输出，可通过 responseConstraint 约束模型输出格式，使 JSON 结构保持稳定68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/789722API
curl https://kongchang.com/api/v1/knowledge/claims/789722MCP
get_claim(id=789722)