GPT 4o support?

[Unlok-Simon]

Any thoughts on when we will be able to select GPT 4o as the model for the OpenAI API? Keen to be able to use it from Jan for performance and cost reasons.

[khandar-william]

Since I saw the configuration exists: https://github.com/janhq/jan/blob/dev/extensions/inference-openai-extension/resources/models.json
Currently I add it manually:

1. in /jan/models create new directory gpt-4o-custom (or any name)
2. create one file model.json
3. fill model.json with

{
    "sources": [
      {
        "url": "https://openai.com"
      }
    ],
    "id": "gpt-4o",
    "object": "model",
    "name": "OpenAI GPT 4o",
    "version": "1.1",
    "description": "OpenAI GPT 4o is a new flagship model with fast speed and high quality",
    "format": "api",
    "settings": {},
    "parameters": {
      "max_tokens": 4096,
      "temperature": 0.7,
      "top_p": 0.95,
      "stream": true,
      "stop": [],
      "frequency_penalty": 0,
      "presence_penalty": 0
    },
    "metadata": {
      "author": "OpenAI",
      "tags": [
        "General"
      ]
    },
    "engine": "openai"
  }

(taken from here)

[Unlok-Simon]

Saw this has been added in the latest release 0.4.13. Thanks!