产品简介

模型列表

为确保API的顺利购买与使用，建议用户提前完成企业账户注册。如果尚未注册，可点击 

 进行快速注册。

提示

用户登录企业账户后，进入[产品/大模型推理服务]页面，点击立即开通API服务按钮，即可快速开启大模型推理之旅，体验高效、智能的服务能力。

目前系统支持DeepSeek-R1满血版模型(FP8无量化版)，用户根据不同的业务需求调用对应的模型。

DeepSeek版本	参数规模（B）	模型大小（约）	上下文长度	最大思维链长度（1）	最大输出长度（2）	model
DeepSeek-R1	671	FP8: 671GB	64K	32K	8K	deepseek-r1

（1） 最大思维链长度：是推理完整性与计算效率的关键参数，用户需根据具体任务和模型能力进行调整。

（2）最大输出长度：指模型生成的回复文本的最大字符数或令牌（Tokens）数量。

用户获取API Key后可直接调用相应的模型进行任务执行。例如：使用如下方式在本地调用满血版的deepseek-r1。

cURL

curl --location 'https://deepseek.alayanew.com/v1/chat/completions' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer <your_access_token>' \
--data ' 
{
    "stream": false,
    "messages": [
        {
            "role": "user",
            "content": "Please generate an essay"
        }
    ],
    "model": "deepseek-r1"
}
'

Python

from openai import OpenAI

client = OpenAI(api_key="<your_access_token>", base_url="https://deepseek.alayanew.com/v1")

response = client.chat.completions.create(
    model="deepseek-r1",
    messages=[
        {"role": "system", "content": "You are a helpful assistant"},
        {"role": "user", "content": "Hello"},
  ],
    max_tokens=1024,
    temperature=0.7,
    stream=False
)

print(response.choices[0].message.content)

GO

package main

import (
        "fmt"
        "io/ioutil"
        "net/http"
        "strings"
)

func main() {

        url := "https://deepseek.alayanew.com/v1/chat/completions"
        method := "POST"

        payload := strings.NewReader(`{
  "messages": [
    {
      "content": "You are a helpful assistant",
      "role": "system"
    },
    {
      "content": "Hi",
      "role": "user"
    }
  ],
  "model": "deepseek-r1",
  "frequency_penalty": 0,
  "max_tokens": 2048,
  "presence_penalty": 0,
  "response_format": {
    "type": "text"
  },
  "stop": null,
  "stream": false,
  "stream_options": null,
  "temperature": 1,
  "top_p": 1,
  "tools": null,
  "tool_choice": "none",
  "logprobs": false,
  "top_logprobs": null
}`)

        client := &http.Client{}
        req, err := http.NewRequest(method, url, payload)

        if err != nil {
                fmt.Println(err)
                return
        }
        req.Header.Add("Content-Type", "application/json")
        req.Header.Add("Accept", "application/json")
        req.Header.Add("Authorization", "Bearer <your_access_token>")

        res, err := client.Do(req)
        if err != nil {
                fmt.Println(err)
                return
        }
        defer res.Body.Close()

        body, err := ioutil.ReadAll(res.Body)
        if err != nil {
                fmt.Println(err)
                return
        }
        fmt.Println(string(body))
}

JavaScript(react为例)

axios

const axios = require('axios');
let data = JSON.stringify({
  "messages": [
    {
      "content": "You are a helpful assistant",
      "role": "system"
    },
    {
      "content": "Hi",
      "role": "user"
    }
  ],
  "model": "deepseek-r1",//模型选择deepseek-r1 / deepseek-r1-lite
  "frequency_penalty": 0,
  "max_tokens": 2048,
  "response_format": {
    "type": "text"
  },
  "stop": null,
  "stream": true,//使用流式传输
  "temperature": 1,
  "logprobs": false,
});

let config = {
  method: 'post',
  maxBodyLength: Infinity,
  url: 'https://deepseek.alayanew.com/v1/chat/completions',//前端请注意跨域请求问题
  headers: {
    'Content-Type': 'application/json',
    'Accept': 'application/json',
    'Authorization': 'Bearer <your_access_token>'
  },
  data: data
};

axios(config)
  .then((response) => {
    console.log("返回信息", JSON.stringify(response.data));
    
  })
  .catch((error) => {
    console.log("错误信息", error);

  });

fetch

import React,{useState} from 'react'
const [requestData,setRequestData]=useState('')
let data = JSON.stringify({
   "messages": [
     {
       "content": "You are a helpful assistant",
       "role": "system"
     },
     {
       "content": "Hi",
       "role": "user"
     }
   ],
   "model": "deepseek-r1",//模型选择deepseek-r1 / deepseek-r1-lite
   "frequency_penalty": 0,
   "max_tokens": 2048,
   "response_format": {
     "type": "text"
   },
   "stop": null,
   "stream": true,//可以自由选择是否为流式传输
   "temperature": 1,
   "logprobs": false,
 });
try {
const response = await fetch('https://deepseek.alayanew.com/v1/chat/completions', {
 method: 'POST',
 headers: {
   'Content-Type': 'application/json',
   'Authorization': 'Bearer <your_access_token>',//请在此处修改为你的API Key
 },
 body: JSON.stringify(data),
});

// 处理流式响应
const reader = response.body.getReader();
const decoder = new TextDecoder('utf-8');
let partialData = '';
const dataStream = data.stream

while (true) {
 const { done, value } = await reader.read();
 if (done) {
   break;
 }
 partialData += decoder.decode(value, { stream: true });
 if (dataStream) {//判断是否为流式响应
   const lines = partialData.split('\n');
   for (let line of lines) {
     if (line.startsWith('data: ')) {
       line = line.substring(6); 
       if (line === '[DONE]') {
         break; // 结束流式响应
       }
       try {
         const jsonData = JSON.parse(line);
         const content = jsonData.choices[0]?.delta?.content;
         if (content) {
           console.log(content)//可以打印在控制台
           setRequestData(prevData => prevData + content);//将消息存储在state中
         }
       } catch (e) {
         console.error('Error parsing JSON:', e);
       }
     }
   }
   // 清空 partialData 以便处理下一个分块
   partialData = '';
 } else {
   const jsonData = JSON.parse(partialData)
   const content = jsonData.choices[0]?.message?.content ?? null;
   if (content !== null) {
     console.log(content);//可以打印在控制台
     setRequestData(content);//将消息存储在state中
   }
 }

}
} catch (error) {
console.error('Final error:', error);
}

Java

import okhttp3.*;

import java.io.IOException;
import java.util.concurrent.TimeUnit;

public class Test {


    private static final String OPENAI_API_KEY = "<your_access_token>";
    private static final String OPENAI_API_BASE = "https://deepseek.alayanew.com/v1/chat/completions";


    public static void main(String[] args) {

        OkHttpClient client = new OkHttpClient().newBuilder()
                .connectTimeout(10, TimeUnit.SECONDS)
                .readTimeout(10,TimeUnit.MINUTES)
                .writeTimeout(10,TimeUnit.MINUTES)
                .build();
        MediaType mediaType = MediaType.parse("application/json");
        RequestBody body = RequestBody.create(mediaType, "{\n  \"messages\": [\n    {\n      \"content\": \"You are a helpful assistant\",\n      \"role\": \"system\"\n    },\n    {\n      \"content\": \"你是一个特级中学语文老师，特别擅长唐诗的研究，会高度模仿李白、杜甫的作诗风格，还特别能接受别人的意见。\",\n      \"role\": \"user\"\n    }\n  ],\n  \"model\": \"deepseek-r1\",\n  \"frequency_penalty\": 0,\n  \"max_tokens\": 2048,\n  \"presence_penalty\": 0,\n  \"response_format\": {\n    \"type\": \"text\"\n  },\n  \"stop\": null,\n  \"stream\": false,\n  \"stream_options\": null,\n  \"temperature\": 1,\n  \"top_p\": 1,\n  \"tools\": null,\n  \"tool_choice\": \"none\",\n  \"logprobs\": false,\n  \"top_logprobs\": null\n}");
        Request request = new Request.Builder()
                .url(OPENAI_API_BASE)
                .method("POST", body)
                .addHeader("Content-Type", "application/json")
                .addHeader("Accept", "application/json")
                .addHeader("Authorization", "Bearer " + OPENAI_API_KEY)
                .build();
        try {
            Response response = client.newCall(request).execute();

            System.out.println(response.body().string());
        }catch (IOException e){
            e.printStackTrace();
        }


    }
}

功能使用

推理服务通过API调用大模型，基本流程如下所示。

• 获取访问令牌
• 发起API请求
• 第三方调用

计费说明

平台提供专属资源部署模型，详情用户可查看计费说明。