Models

OpenGPT0 model IDs use the format provider/model-id. The default product keeps this list short:

• browser/chatgpt for ChatGPT through the Browser provider.
• ollama/qwen3.5:9b-mlx for the built-in Ollama local model.
• rapid-mlx/qwen3.5-9b-4bit for the built-in Rapid MLX local model.

Default Model

For the product experience, use Browser:

opengpt0.json

{
  "$schema": "https://opengpt0.com/config.json",
  "model": "browser/chatgpt"
}

Browser Variants

Browser exposes ChatGPT intelligence variants through the same browser/chatgpt model:

• instant
• medium
• high
• extra-high
• pro

Local Models

For local usage, install Ollama, pull the built-in model, then select the matching OpenGPT0 model ID:

ollama pull qwen3.5:9b-mlx

opengpt0.json

{
  "$schema": "https://opengpt0.com/config.json",
  "model": "ollama/qwen3.5:9b-mlx"
}

For Rapid MLX, point the local runtime at http://127.0.0.1:8080/v1 and select:

opengpt0.json

{
  "$schema": "https://opengpt0.com/config.json",
  "model": "rapid-mlx/qwen3.5-9b-4bit"
}

Only Browser, Ollama, and Rapid MLX models are built in. Keeping the model list narrow makes the product easier to test and keeps provider behavior explicit.