scalaone / azure-openai-proxy Goto Github PK

A tool that transforms OpenAI API requests into Azure OpenAI API requests, allowing OpenAI-compatible applications to seamlessly use Azure OpenAI. 一个 OpenAI API 的代理工具，能将 OpenAI API 请求转为 Azure OpenAI API 请求，从而让只支持 OpenAI 的应用程序无缝使用 Azure OpenAI。

Home Page: https://gptlite.vercel.app

License: MIT License

JavaScript 0.33% TypeScript 96.27% Dockerfile 3.41%

azure-openai azure-openai-proxy azureopenai chatgpt nextjs nextjs13 gpt-4 gpt-4-32k gpt-35-turbo

azure-openai-proxy's Introduction

Azure OpenAI Proxy

English | 简体中文

Azure OpenAI Proxy is a tool that transforms OpenAI API requests into Azure OpenAI API requests, allowing OpenAI-compatible applications to seamlessly use Azure Open AI.

Prerequisites

An Azure OpenAI account is required to use Azure OpenAI Proxy.

Azure Deployment

Remember to:

Select the region that matches your Azure OpenAI resource for best performance.
If deployment fails because the 'proxywebapp' name is already taken, change the resource prefix and redeploy.
The deployed proxy app is part of a B1 pricing tier Azure web app plan, which can be modified in the Azure Portal after deployment.

Docker Deployment

To deploy using Docker, execute the following command:

docker run -d -p 3000:3000 scalaone/azure-openai-proxy

Local Execution and Testing

Follow these steps:

Install NodeJS 20.
Clone the repository in the command line window.
Run npm install to install the dependencies.
Run npm start to start the application.
Use the script below for testing. Replace AZURE_RESOURCE_ID, AZURE_MODEL_DEPLOYMENT, and AZURE_API_KEY before running. The default value for AZURE_API_VERSION is 2024-02-01 and is optional.

Test script

```bash curl -X "POST" "http://localhost:3000/v1/chat/completions" \ -H 'Authorization: AZURE_RESOURCE_ID:AZURE_MODEL_DEPLOYMENT:AZURE_API_KEY:AZURE_API_VERSION' \ -H 'Content-Type: application/json; charset=utf-8' \ -d $'{ "messages": [ { "role": "system", "content": "You are an AI assistant that helps people find information." }, { "role": "user", "content": "hi." } ], "temperature": 1, "model": "gpt-3.5-turbo", "stream": false }' ```

Tested Applications

The azure-openai-proxy has been tested and confirmed to work with the following applications:

Application Name	Docker-compose File for E2E Test
chatgpt-lite	docker-compose.yml
chatgpt-minimal	docker-compose.yml
chatgpt-next-web	docker-compose.yml
chatbot-ui	docker-compose.yml
chatgpt-web	docker-compose.yml

To test locally, follow these steps:

Clone the repository in a command-line window.
Update the OPENAI_API_KEY environment variable with AZURE_RESOURCE_ID:AZURE_MODEL_DEPLOYMENT:AZURE_API_KEY. Alternatively, update the OPENAI_API_KEY value in the docker-compose.yml file directly.
Navigate to the directory containing the docker-compose.yml file for the application you want to test.
Execute the build command: docker-compose build.
Start the service: docker-compose up -d.
Access the application locally using the port defined in the docker-compose.yml file. For example, visit http://localhost:3000.

FAQs

Q: What are `AZURE_RESOURCE_ID`,`AZURE_MODEL_DEPLOYMENT`, and `AZURE_API_KEY`?

A: These can be found in the Azure management portal. See the image below for reference: ![resource-and-model](./docs/images/resource-and-model.jpg)

Q: How can I use gpt-4 and gpt-4-32k models?

A: To use gpt-4 and gpt-4-32k models, follow the key format below: `AZURE_RESOURCE_ID:gpt-3.5-turbo|AZURE_MODEL_DEPLOYMENT,gpt-4|AZURE_MODEL_DEPLOYMENT,gpt-4-32k|AZURE_MODEL_DEPLOYMENT:AZURE_API_KEY:AZURE_API_VERSION`

Contributing

We welcome all PR submissions.

azure-openai-proxy's People

Contributors

Stargazers

Watchers

azure-openai-proxy's Issues

一键部署到Azure出现错误Website with given name proxywebapp already exists

proxyappplan可以部署，proxywebapp部署错误

Enable proxy support for the Embeddings API

trouble using stream mode

I've trouble using stream mode. Could you please provide an example client that demonstrates how to use Stream mode effectively?

Dockerfile build Error

 => CACHED [deps 3/6] RUN apk add --no-cache libc6-compat                                                                                                                                                       0.0s
 => CACHED [deps 4/6] WORKDIR /app                                                                                                                                                                              0.0s
 => CACHED [deps 5/6] COPY package.json package-lock.json* ./                                                                                                                                                   0.0s
 => ERROR [deps 6/6] RUN npm ci                                                                                                                                                                                 0.2s
------
 > [deps 6/6] RUN npm ci:
------
Dockerfile:15
--------------------
  13 |     # Install dependencies based on the preferred package manager
  14 |     COPY package.json package-lock.json* ./
  15 | >>> RUN npm ci
  16 |     
  17 |     # Rebuild the source code only when needed
--------------------
ERROR: failed to solve: process "/bin/sh -c npm ci" did not complete successfully: exit code: 139
ERROR: Service 'azure-openai-proxy' failed to build : Build failed

Intermittent 'ECONNRESET' Errors

Reviewing the past month's log in my deployment, I've noticed intermittent occurrences of "Error in getCompletions: Error: read ECONNRESET" from client apps. This error typically arises when the connection between the proxy and the Azure OpenAI server unexpectedly closes during request processing. The "ECONNRESET" message signifies that the server abruptly closed the connection.

To address this, the code should be enhanced to handle such situations appropriately. This could involve adding a retry mechanism for network issues or implementing a timeout.

Azure GPT-4 model overloaded requests error

When invoking Azure GPT 4 models, you may encounter an error message stating That model is currently overloaded with other requests. You can retry your request, or contact us through an Azure support request. In such cases, the proxy should automatically retry.