Rerank

POST/v1/rerank

根据查询对候选文档重新排序。

结果结构和相关性分数由所选模型 Provider 返回。本文档不假设固定分数范围或批量上限。

核心请求字段

字段	类型	必填	说明
`model`	string	是	Rerank 模型 ID。
`query`	string	是	用于相关性排序的查询文本。
`documents`	array	是	需要排序的候选文档。
`top_n`	integer	否	期望返回的结果数量，仅在模型支持时使用。

请求示例

curl https://api.noxgate.net/v1/rerank \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{"model":"YOUR_MODEL_ID","query":"What is NoxGate?","documents":["An AI API gateway","A weather report"],"top_n":2}'

import requests

response = requests.post(
    "https://api.noxgate.net/v1/rerank",
    headers={"Authorization": "Bearer sk-your-api-key"},
    json={"model": "YOUR_MODEL_ID", "query": "What is NoxGate?", "documents": ["An AI API gateway", "A weather report"]},
)
print(response.json())

const response = await fetch("https://api.noxgate.net/v1/rerank", {
  method: "POST",
  headers: { Authorization: "Bearer sk-your-api-key", "Content-Type": "application/json" },
  body: JSON.stringify({ model: "YOUR_MODEL_ID", query: "What is NoxGate?", documents: ["An AI API gateway", "A weather report"] }),
});
console.log(await response.json());

成功响应

{"id":"rerank_example","results":[{"index":0,"relevance_score":0.97},{"index":1,"relevance_score":0.08}],"usage":{"total_tokens":18}}