Quick Start

3 分钟快速接入

First time using an API platform? Follow the four steps below: pay, check what your order returned, then either use the API Key directly or redeem the code and create your own key.

第一次接 API 也没关系。你只要按下面 4 步走：先支付，再看订单实际返回了什么，然后要么直接用 API Key，要么先兑换再创建自己的 Key。

Step 1  —  Buy a package 第一步  —  购买套餐

Head to the pricing section and pick a package that fits your needs. After payment, keep your order ID. The success page and Order Lookup page are the two places where AiCredits will show your delivery result.

前往套餐页选择适合你的档位。支付完成后先保留订单号；AiCredits 会在成功页和 订单查询 页展示你的交付结果。

Step 2  —  Get the delivery result 第二步  —  获取交付结果

After payment, the most important thing is to identify which result your order returned. Do not go to the console first. The success page and Order Lookup page are the source of truth.

支付完成后，最重要的是先看清楚这笔订单到底返回了什么。不要一上来就进控制台。成功页和 订单查询 页才是当前这笔订单的准确信息来源。

Step 3  —  Use the direct API or fallback flow 第三步  —  直接使用 API 或走兜底兑换

There are only two real situations after payment. Match what you see on the page with one of the two boxes below and follow that exact path.

支付之后其实只有两种真实情况。你只要把页面上看到的内容，对照下面两种情况走就行，不要自己脑补第三种流程。

The two values you eventually need in your project always look like this:

不管走哪条路，最后你放进项目里的都只有这两项：

API Base URL:  http://168.144.39.53:3000/v1
API Key:      sk-<your-key-from-order-page-or-console>

Step 4  —  Make your first call 第四步  —  发起第一次调用

Pick your language, then replace only two placeholders: api_key and base_url. For your first test, keep the model as deepseek-chat.

选一个你熟悉的语言，然后只替换两个地方：api_key 和 base_url。第一次测试模型先保持 deepseek-chat 不要改。

from openai import OpenAI

client = OpenAI(
    api_key="sk-<your-key>",
    base_url="http://168.144.39.53:3000/v1",
)


response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"},
    ],
)
print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-<your-key>",
  baseURL: "http://168.144.39.53:3000/v1",
});


const response = await client.chat.completions.create({
  model: "deepseek-chat",
  messages: [
    { role: "system", content: "You are a helpful assistant." },
    { role: "user",   content: "Hello!" },
  ],
});

console.log(response.choices[0].message.content);

curl http://168.144.39.53:3000/v1/chat/completions \
  -H "Authorization: Bearer sk-<your-key>" \

  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user",   "content": "Hello!"}
    ]
  }'

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.chat.*;

OpenAIClient client = OpenAIOkHttpClient.builder()
    .apiKey("sk-<your-key>")
    .baseUrl("http://168.144.39.53:3000/v1")
    .build();


ChatCompletion completion = client.chat().completions().create(
    ChatCompletionCreateParams.builder()
        .model("deepseek-chat")
        .addUserMessage("Hello!")
        .build()
);
System.out.println(completion.choices().get(0).message().content());

package main

import (
    "context"
    "fmt"
    openai "github.com/openai/openai-go"
    "github.com/openai/openai-go/option"
)

func main() {
    client := openai.NewClient(
        option.WithAPIKey("sk-<your-key>"),
        option.WithBaseURL("http://168.144.39.53:3000/v1"),
    )

    resp, _ := client.Chat.Completions.New(context.Background(),
        openai.ChatCompletionNewParams{
            Model: openai.F("deepseek-chat"),
            Messages: openai.F([]openai.ChatCompletionMessageParamUnion{
                openai.UserMessage("Hello!"),
            }),
        },
    )
    fmt.Println(resp.Choices[0].Message.Content)
}

Supported Models 支持的模型

Pass any of the model IDs below as the model parameter.

将下面的模型 ID 填入 model 参数即可。

FAQ

Do I need to create an account?

需要注册账号吗？

Usually no. In the fast path, AiCredits gives you a ready-to-use API Key directly after payment, so you can start calling immediately. You only need an account if that order falls back to redemption-code delivery and you choose to redeem it manually in the console.

通常不需要。快速路径下，AiCredits 支付后会直接给你可用 API Key，你可以立刻调用。只有当该订单回退成兑换码交付时，你才需要去控制台注册账号并手动兑换。

Does quota expire?

额度会过期吗？

No. Credits are prepaid and do not expire unless otherwise stated on the pricing page. Use them at your own pace.

不会。额度是预付制，除定价页特别说明外永久有效，用多少是多少。

How do I check my remaining balance?

怎么查剩余额度？

Log into the AiCredits console. The dashboard shows your current quota, usage history, and token consumption per model.

登录控制台，首页仪表盘会显示当前剩余额度、使用历史和每个模型的 token 消耗情况。

What if I get a 401 Unauthorized error?

遇到 401 Unauthorized 怎么办？

Double-check that your Authorization header is exactly Bearer sk-<your-key> and that the base_url points to http://168.144.39.53:3000/v1 (not the OpenAI endpoint).

检查 Authorization 头是否为 Bearer sk-<你的key>， 以及 base_url 是否指向 http://168.144.39.53:3000/v1（不是 OpenAI 的地址）。

Can I use streaming?

支持流式输出吗？

Yes. Set stream: true in your request — it works exactly like OpenAI streaming.

支持。请求中加上 stream: true，行为与 OpenAI 完全一致。

I lost my redemption code. What now?

兑换码丢了怎么办？

Go to Order Lookup, enter your order ID (saved in your browser or sent by email), and you'll see your code again.

去订单查询页，输入订单号（浏览器有保存，或查看邮件），即可找回兑换码。

Still stuck?

还有问题？

Email us at mrwangyanlong2026@outlook.com — we reply within 24 hours.

发邮件到 mrwangyanlong2026@outlook.com，24 小时内回复。