A lightweight gRPC + RESTful wrapper for AI models. Built in Go, this service exposes a unified interface for AI generation via both protocol-level gRPC and HTTP/JSON endpoints. Includes Swagger UI, Docker support, and flexible configuration via flags or environment variables.

• ✅ gRPC service: AiService.GenAi(prompt)
• ✅ RESTful HTTP endpoint: POST /v1/genai
• ✅ Swagger UI for interactive API testing
• ✅ Docker-ready with multi-port support
• ✅ Configurable via flags, .env, or inline environment variables
• ✅ Graceful shutdown and signal handling
• ✅ Gemini API key integration

go run main.go --port=8000 --api-key=your_key_here --model=gemini-2.5-pro

Or use environment variables:

export PORT=8000
export API_KEY=your_key_here
export MODEL=gemini-2.5-pro
go run main.go

docker build -t wrapper .

Or pull from GitHub Container Registry:

docker pull ghcr.io/imrany/wrapper

docker run -d --name wrapper \
  -p 8000:8000 \  # gRPC
  -p 8090:8090 \  # REST + Swagger
  ghcr.io/imrany/wrapper \
  --port=8000 \
  --api-key=your_key_here
  --model=gemini-2.5-pro

Create .env:

PORT=8000
API_KEY="your_key_here"
MODEL="gemini-2.5-pro" # or "gpt-5.1-2025-11-13"

Run:

docker run --env-file .env -p 8000:8000 -p 8090:8090 ghcr.io/imrany/wrapper

POST /v1/genai
Content-Type: application/json
Body: "Hello AI"

curl -X POST http://localhost:8090/v1/genai \
  -H "Content-Type: application/json" \
  -d '"Hello AI"'

service AiService {
  rpc GenAi(GenAiRequest) returns (GenAiResponse);
}

grpcurl -insecure localhost:8000 \
  wekalist.api.v1.AiService.GenAi \
  -d '{"prompt": "Hello AI"}'

Visit: http://localhost:8090/swagger/

{
  "prompt": "Hello AI",
  "response": "Hello! How can I help you today?\n",
  "status": null
}

GET /healthz

Returns 200 OK with body ok.

Built by Imran Licensed under MIT