Deploy an OpenAI Proxy on Alibaba Cloud Function Compute in Minutes

Introduction

Since OpenAI released ChatGPT in November 2022, the platform has evolved with OpenAPI, GPT‑4, and ChatGPT Plugins, positioning ChatGPT as an operating‑system‑level entry point that spurs new applications across industries.

Developers need a reliable overseas endpoint for OpenAI API calls to avoid IP conflicts and account bans during local debugging or server deployment.

Solution Overview

The article presents a practical solution using Alibaba Cloud Function Compute (FC), which offers generous free trial quotas, to host a stable proxy for OpenAI API requests.

Command‑Line Deployment

Preparation

Open the Alibaba Cloud console and enable Function Compute FC ( link ).

Install the latest Node.js version from the official site.

Install Serverless Devs Tool globally: npm install -g @serverless-devs/s or yarn global add @serverless-devs/s Obtain an Alibaba Cloud AccessKey ID and Secret from the user center.

Configure the Serverless Devs Tool with the AccessKey:

$ s config add
? Please select a provider: Alibaba Cloud (alibaba)
? AccessKeyID (enter your AccessKeyID)
? AccessKeySecret (enter your AccessKeySecret)
Alias: alibaba-access
✔ Configuration successful

Two‑Command Deployment

Step 1: Initialize the project using the command: $ s init openai-proxy Follow the prompts to select a region (e.g., us‑west‑1) and the previously created AccessKey alias.

Step 2: Deploy with a single command : $ s deploy The tool creates the service, function, triggers, and a custom domain. After deployment, note the generated domain name highlighted in the output.

Application Center Deployment (No CLI)

Open the Alibaba Cloud console and enable Function Compute FC ( link ).

Visit the Serverless Devs Application Center page for openai‑proxy ( link ).

Click the “One‑Click Deploy” button.

Select “Create and Deploy Default Environment”.

After a short build, the console shows an “Access Domain” link; this domain is the proxy endpoint.

Using the Proxy

With the proxy URL configured, any OpenAI‑based project can point its OPENAI_API_BASE_URL to the new domain. Example steps for the chatgpt‑demo project:

Clone the repository.

Run npm install to install dependencies.

Copy .env.example to .env, set OPENAI_API_KEY and OPENAI_API_BASE_URL to the proxy domain.

Start the app with npm start and verify it runs at http://localhost:3000/.

Other open‑source ChatGPT front‑ends (e.g., chatbot‑ui , langflow ) can be tested similarly.

Conclusion

Deploying an OpenAI proxy on Alibaba Cloud Function Compute provides a stable overseas call source, eliminating the need for additional networking tricks during local development and testing. The serverless model offers easy deployment, zero‑maintenance operation, and cost efficiency because resources scale to zero when idle.

Because both OpenAI usage and Function Compute are billed on a pay‑as‑you‑go basis, this architecture aligns cost with actual traffic, making it ideal for low‑cost, high‑scalability ChatGPT applications and even broader SaaS scenarios.