Repository files navigation

Gemini-OpenAI-Proxy is a proxy designed to convert the OpenAI API protocol to the Google Gemini protocol. This enables applications built for the OpenAI API to seamlessly communicate with the Gemini protocol, including support for Chat Completion, Embeddings, and Model(s) endpoints.

• Gemini-OpenAI-Proxy
  • Table of Contents
  • Build
  • Deploy
  • Usage
  • Compatibility
  • License

To build the Gemini-OpenAI-Proxy, follow these steps:

go build -o gemini main.go

We recommend deploying Gemini-OpenAI-Proxy using Docker for a straightforward setup. Follow these steps to deploy with Docker:

You can either do this on the command line:

docker run --restart=unless-stopped -it -d -p 8080:8080 --name gemini zhu327/gemini-openai-proxy:latest

Or with the following docker-compose config:

version:'3'services:gemini:container_name:geminienvironment:# Set Environment Variables here. Defaults listed below         -GPT_4_VISION_PREVIEW=gemini-1.5-flash-latest         -DISABLE_MODEL_MAPPING=0ports:         -"8080:8080"image:zhu327/gemini-openai-proxy:latestrestart:unless-stopped

Adjust the port mapping (e.g.,-p 8080:8080) as needed, and ensure that the Docker image version (zhu327/gemini-openai-proxy:latest) aligns with your requirements.

Gemini-OpenAI-Proxy offers a straightforward way to integrate OpenAI functionalities into any application that supports custom OpenAI API endpoints. Follow these steps to leverage the capabilities of this proxy:

1. Set Up OpenAI Endpoint:Ensure your application is configured to use a custom OpenAI API endpoint. Gemini-OpenAI-Proxy seamlessly works with any OpenAI-compatible endpoint.

2. Get Google AI Studio API Key:Before using the proxy, you'll need to obtain an API key fromai.google.dev. Treat this API key as your OpenAI API key when interacting with Gemini-OpenAI-Proxy.

3. Integrate the Proxy into Your Application:Modify your application's API requests to target the Gemini-OpenAI-Proxy, providing the acquired Google AI Studio API key as if it were your OpenAI API key.

   Example Chat Completion API Request (Assuming the proxy is hosted athttp://localhost:8080):

   curl http://localhost:8080/v1/chat/completions \ -H"Content-Type: application/json" \ -H"Authorization: Bearer$YOUR_GOOGLE_AI_STUDIO_API_KEY" \ -d'{     "model": "gpt-3.5-turbo",     "messages": [{"role": "user", "content": "Say this is a test!"}],     "temperature": 0.7 }'

   Alternatively, use Gemini Pro Vision:

   curl http://localhost:8080/v1/chat/completions \ -H"Content-Type: application/json" \ -H"Authorization: Bearer$YOUR_GOOGLE_AI_STUDIO_API_KEY" \ -d'{     "model": "gpt-4-vision-preview",     "messages": [{"role": "user", "content": [        {"type": "text", "text": "What’s in this image?"},        {          "type": "image_url",          "image_url": {            "url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg"          }        }     ]}],     "temperature": 0.7 }'

   If you wish to mapgpt-4-vision-preview togemini-1.5-pro-latest, you can configure the environment variableGPT_4_VISION_PREVIEW = gemini-1.5-pro-latest. This is becausegemini-1.5-pro-latest now also supports multi-modal data. Otherwise, the default uses thegemini-1.5-flash-latest model

   If you already have access to the Gemini 1.5 Pro api, you can use:

   curl http://localhost:8080/v1/chat/completions \ -H"Content-Type: application/json" \ -H"Authorization: Bearer$YOUR_GOOGLE_AI_STUDIO_API_KEY" \ -d'{     "model": "gpt-4-turbo-preview",     "messages": [{"role": "user", "content": "Say this is a test!"}],     "temperature": 0.7 }'

   Example Embeddings API Request:

   curl http://localhost:8080/v1/embeddings \ -H"Content-Type: application/json" \ -H"Authorization: Bearer$YOUR_GOOGLE_AI_STUDIO_API_KEY" \ -d'{    "model": "text-embedding-ada-002",    "input": "This is a test sentence." }'

   You can also pass in multiple input strings as a list:

   curl http://localhost:8080/v1/embeddings \ -H"Content-Type: application/json" \ -H"Authorization: Bearer$YOUR_GOOGLE_AI_STUDIO_API_KEY" \ -d'{    "model": "text-embedding-ada-002",    "input": ["This is a test sentence.", "This is another test sentence"] }'

   Model Mapping:

   GPT Model	Gemini Model
   gpt-4	gemini-1.5-flash-002
   gpt-4-turbo-preview	gemini-1.5-pro-latest
   gpt-4o	gemini-2.0-flash-exp
   text-embedding-ada-002	text-embedding-004

   If you want to disable model mapping, configure the environment variableDISABLE_MODEL_MAPPING=1. This will allow you to refer to the Gemini models directly.

   Here is an example API request with model mapping disabled:

   curl http://localhost:8080/v1/chat/completions \ -H"Content-Type: application/json" \ -H"Authorization: Bearer$YOUR_GOOGLE_AI_STUDIO_API_KEY" \ -d'{     "model": "gemini-1.0-pro-latest",     "messages": [{"role": "user", "content": "Say this is a test!"}],     "temperature": 0.7 }'

4. Handle Responses:Process the responses from the Gemini-OpenAI-Proxy in the same way you would handle responses from OpenAI.

Now, your application is equipped to leverage OpenAI functionality through the Gemini-OpenAI-Proxy, bridging the gap between OpenAI and applications using the Google Gemini Pro protocol.

• #4

Google AI Studio now provides an OpenAI-compatible API endpoint. Users of this project are recommended to use the officially provided compatibility API. Documentation is available at:

https://ai.google.dev/gemini-api/docs/openai

Gemini-OpenAI-Proxy is licensed under the MIT License - see theLICENSE file for details.