Responses

OpenAI Responses API 接口。Sakrylle 网关把它当作 OpenAI 平台原生入口。

协议边界

这是 OpenAI Responses 格式。Anthropic 上游分组可能通过网关转换接入，但如果你需要 Anthropic 原生工具或内容块语义，优先使用 /v1/messages。

兼容性提示

Sakrylle 平台默认通过此端点调度 OpenAI 通道；当 API Key 所在分组被管理员标记为「仅限 Claude Code」时，访问 /v1/responses 会被拒绝。常规用法仍推荐使用 /v1/chat/completions，参数集更熟悉。

Endpoint

POST https://api.sakrylle.com/v1/responses

请求头

名称	必填	说明
Authorization	是	Bearer sk-xxxxxxxxxxxxxxxx
Content-Type	是	application/json
Accept	否	流式可写 text/event-stream

请求体参数

字段	类型	必填	说明
model	string	是	OpenAI 模型 ID，例如 gpt-5.6-sol
input	string | array	是	用户输入，可为字符串或消息/内容数组
instructions	string	否	系统级指令，等价于 chat 的 system
stream	boolean	否	默认 false；为 true 时返回 SSE 事件流
temperature	number	否	0–2
top_p	number	否	核采样概率
max_output_tokens	integer	否	最大输出 token 数
tools	array	否	工具/函数定义
tool_choice	string | object	否	auto / none / 指定工具
metadata	object	否	任意 KV，回写到响应中
previous_response_id	string	否	续接历史 response

网关只校验 model 必填、JSON 合法，其余字段透传。

请求体示例

{
  "model": "gpt-5.6-sol",
  "instructions": "你是一个简洁的中文助手。",
  "input": "自我介绍一下。"
}

响应示例

{
  "id": "resp_xxxxxxxxxxxxxxxx",
  "object": "response",
  "created_at": 1716350400,
  "model": "gpt-5.6-sol",
  "status": "completed",
  "output": [
    {
      "id": "msg_xxxxxxxxxxxxxxxx",
      "type": "message",
      "role": "assistant",
      "content": [
        {
          "type": "output_text",
          "text": "我是 gpt-5.6-sol，有什么可以帮您？"
        }
      ]
    }
  ],
  "usage": {
    "input_tokens": 22,
    "output_tokens": 14,
    "total_tokens": 36
  }
}

流式响应示例

stream: true 后返回 SSE，事件类型例如 response.created、response.output_text.delta、response.completed：

event: response.created
data: {"type":"response.created","response":{"id":"resp_xxx","model":"gpt-5.6-sol","status":"in_progress"}}

event: response.output_text.delta
data: {"type":"response.output_text.delta","delta":"Sakrylle"}

event: response.output_text.delta
data: {"type":"response.output_text.delta","delta":" API"}

event: response.completed
data: {"type":"response.completed","response":{"id":"resp_xxx","status":"completed","usage":{"input_tokens":22,"output_tokens":14,"total_tokens":36}}}

错误

• 400 invalid_request_error — 缺 model 或请求体非 JSON
• 401 authentication_error — API Key 无效
• 403 permission_error — 分组被限制为「仅限 Claude Code」时拒绝 /v1/responses
• 403 billing_error — 余额或订阅限额不足
• 429 rate_limit_error — 并发或速率限制
• 502 upstream_error / server_error — 上游连续失败，请稍后重试

更多错误码见 错误处理。

代码示例

curl

curl https://api.sakrylle.com/v1/responses \
  -H "Authorization: Bearer sk-xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.6-sol",
    "input": "Hello!"
  }'

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-xxxxxxxxxxxxxxxx",
    base_url="https://api.sakrylle.com/v1",
)

resp = client.responses.create(
    model="gpt-5.6-sol",
    input="Hello!",
)
print(resp.output_text)

Node

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-xxxxxxxxxxxxxxxx",
  baseURL: "https://api.sakrylle.com/v1",
});

const resp = await client.responses.create({
  model: "gpt-5.6-sol",
  input: "Hello!",
});
console.log(resp.output_text);

Go

package main

import (
    "bytes"
    "io"
    "net/http"
    "fmt"
)

func main() {
    body := []byte(`{"model":"gpt-5.6-sol","input":"Hello!"}`)
    req, _ := http.NewRequest("POST",
        "https://api.sakrylle.com/v1/responses",
        bytes.NewReader(body))
    req.Header.Set("Authorization", "Bearer sk-xxxxxxxxxxxxxxxx")
    req.Header.Set("Content-Type", "application/json")
    resp, err := http.DefaultClient.Do(req)
    if err != nil {
        panic(err)
    }
    defer resp.Body.Close()
    out, _ := io.ReadAll(resp.Body)
    fmt.Println(string(out))
}

Rust

use reqwest::Client;
use serde_json::json;

#[tokio::main]
async fn main() -> anyhow::Result<()> {
    let client = Client::new();
    let resp: serde_json::Value = client
        .post("https://api.sakrylle.com/v1/responses")
        .bearer_auth("sk-xxxxxxxxxxxxxxxx")
        .json(&json!({
            "model": "gpt-5.6-sol",
            "input": "Hello!"
        }))
        .send()
        .await?
        .json()
        .await?;
    println!("{resp:#}");
    Ok(())
}

Java

import java.net.URI;
import java.net.http.*;

public class Responses {
  public static void main(String[] args) throws Exception {
    String body = """
      {"model":"gpt-5.6-sol","input":"Hello!"}
      """;
    HttpRequest req = HttpRequest.newBuilder()
      .uri(URI.create("https://api.sakrylle.com/v1/responses"))
      .header("Authorization", "Bearer sk-xxxxxxxxxxxxxxxx")
      .header("Content-Type", "application/json")
      .POST(HttpRequest.BodyPublishers.ofString(body))
      .build();
    HttpResponse<String> resp = HttpClient.newHttpClient()
      .send(req, HttpResponse.BodyHandlers.ofString());
    System.out.println(resp.body());
  }
}

C#

using System.Net.Http.Headers;
using System.Text;

var http = new HttpClient();
http.DefaultRequestHeaders.Authorization =
    new AuthenticationHeaderValue("Bearer", "sk-xxxxxxxxxxxxxxxx");

var payload = """
  {"model":"gpt-5.6-sol","input":"Hello!"}
  """;
var resp = await http.PostAsync(
    "https://api.sakrylle.com/v1/responses",
    new StringContent(payload, Encoding.UTF8, "application/json"));
Console.WriteLine(await resp.Content.ReadAsStringAsync());