Server to use Enchanted app with OpenAI API

This is a simple server that can be used to interact with the OpenAI API via Ollama API. Made to work with Enchanted.

Why

Enchanted app is a great tool to chatting with self hosted AI models. It supports Ollama API.

OpenAI API is different: it has other endpoints and requests format.

This server will convert a couple of endpoints mimic the Ollama API and converts data to work with OpenAI API.

To do

• Add tests
• Add correct error handling

How to use

Docker

docker run -d \
  --name openai-api-server-for-enchanted \
  -e OPENAI_API_BASE_URL=https://openrouter.ai/api \
  -e OPENAI_ALLOWED_MODELS= \
  -p 3033:3033 \
  --restart unless-stopped \
  ghcr.io/talyguryn/openai-api-server-for-enchanted-app:latest

Docker Compose

version: "3.4"
services:
  openai-api-server-for-enchanted:
    image: ghcr.io/talyguryn/openai-api-server-for-enchanted-app:latest
    container_name: openai-api-server-for-enchanted
    mem_limit: 64m
    environment:
      - OPENAI_API_BASE_URL=https://openrouter.ai/api
      - OPENAI_ALLOWED_MODELS=
    ports:
      - 3033:3033
    restart: unless-stopped

How to develop

Clone the repository and install dependencies:

npm i

If you need to override vars create a .env file as a copy of .env.example and fill required variables.

Run the server:

npm start

ENV variables

You can set the following environment variables:

Env	Description	Default	Example
OPENAI_API_BASE_URL	Base URL for the OpenAI API	https://api.openai.com	https://api.openai.com
OPENAI_ALLOWED_MODELS	Comma separated list of allowed models	None	gpt-3.5-turbo,gpt-4o

All variables are optional.

Server Endpoints

On each request Enchanted app will send Authorization header with the Bearer token. So you can use it to authenticate requests to OpenAI API.

HEAD /

Health check.

Ollama returns 200 with no body.

GET /api/tags

Get list of available models.

Example Ollama response:

{
  "models": [
    {
      "name": "deepseek-r1:7b",
      "model": "deepseek-r1:7b",
      "modified_at": "2025-03-23T07:52:42.164862186Z",
      "size": 4683075271,
      "digest": "0a8c266910232fd3291e71e5ba1e058cc5af9d411192cf88b6d30e92b6e73163",
      "details": {
        "parent_model": "",
        "format": "gguf",
        "family": "qwen2",
        "families": ["qwen2"],
        "parameter_size": "7.6B",
        "quantization_level": "Q4_K_M"
      }
    },
    {
      "name": "llama3.2:3b",
      "model": "llama3.2:3b",
      "modified_at": "2025-03-23T07:52:41.339842922Z",
      "size": 2019393189,
      "digest": "a80c4f17acd55265feec403c7aef86be0c25983ab279d83f3bcd3abbcb5b8b72",
      "details": {
        "parent_model": "",
        "format": "gguf",
        "family": "llama",
        "families": ["llama"],
        "parameter_size": "3.2B",
        "quantization_level": "Q4_K_M"
      }
    }
  ]
}

Example OpenAI response on /v1/models:

{
  "object": "list",
  "data": [
    {
      "id": "gpt-4",
      "object": "model",
      "created": 1687882411,
      "owned_by": "openai"
    },
    {
      "id": "o1-pro",
      "object": "model",
      "created": 1742251791,
      "owned_by": "system"
    }
  ]
}

POST /api/chat

Chat with the model. Streaming allowed.

Request:

{
  "messages": [{ "content": "hey", "images": [], "role": "user" }],
  "model": "deepseek-r1:1.5b",
  "stream": true,
  "options": { "temperature": 0 }
}

Ollama response chunks:

{
  model: "deepseek-r1:1.5b",
  created_at: "2025-03-22T17:01:31.748659Z",
  message: { role: "assistant", content: "\n\n" },
  done: false,
}
{
  model: "deepseek-r1:1.5b",
  created_at: "2025-03-22T17:01:37.788115Z",
  message: { role: "assistant", content: "" },
  done_reason: "stop",
  done: true,
  total_duration: 12801095381,
  load_duration: 4858364600,
  prompt_eval_count: 4,
  prompt_eval_duration: 1511149623,
  eval_count: 41,
  eval_duration: 6429075170,
}

OpenAI response chunks on /v1/chat/completions:

data: {
  id: "chatcmpl-BDwZHr1gDRCmTsBDHgLXmlB90fRhU",
  object: "chat.completion.chunk",
  created: 1742663099,
  model: "gpt-4o-2024-08-06",
  service_tier: "default",
  system_fingerprint: "fp_90d33c15d4",
  choices: [
    {
      index: 0,
      delta: { content: "." },
      logprobs: null,
      finish_reason: null,
    },
  ],
}
data: {
  id: "chatcmpl-BDwZHr1gDRCmTsBDHgLXmlB90fRhU",
  object: "chat.completion.chunk",
  created: 1742663099,
  model: "gpt-4o-2024-08-06",
  service_tier: "default",
  system_fingerprint: "fp_90d33c15d4",
  choices: [{ index: 0, delta: {}, logprobs: null, finish_reason: "stop" }],
}
data: [DONE]