Use inference snaps in OpenCode

OpenCode is an open source AI coding agent. It’s available as a terminal-based interface, desktop app, or IDE extension. This guide will help you configure OpenCode to use an inference snap as a provider for AI-powered features.

You must have OpenCode and an inference snap installed to continue.

Tip

OpenCode is available as a snap. To install the snap, run:

sudo snap install opencode --classic

Get the inference snap endpoint and model name

Use the inference snap’s status command to retrieve the OpenAI endpoint URL and model name.

For example, with the gemma4 snap:

gemma4 status

The output will show the endpoint and model information:

engine: nvidia-gpu
services:
    server: active
    server-webui: active
endpoints:
    openai: http://localhost:8336/v1
    webui: http://localhost:8337/
model:
    name: gemma4-e4b-q4-k-m

Note the openai endpoint URL and the model.name value, as you’ll need them for configuration.

Configure OpenCode

Open or create the OpenCode configuration file at ~/.config/opencode/opencode.jsonc.

Add a new provider configuration with the endpoint and model information:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "gemma4": {
      "api": "openai",
      "name": "Gemma 4 Inference Snap",
      "options": {
        "baseURL": "http://localhost:8336/v1"
      },
      "models": {
        "gemma4-e4b-q4-k-m": {}
      }
    }
  }
}

Update the configuration values to match your specific snap:

  • Provider key (gemma4): A unique identifier for this provider

  • name: A descriptive name for the provider (e.g., “Gemma 4 Inference Snap”)

  • baseURL: The OpenAI endpoint URL from the status command

  • models: An object containing the model name from the status command

Use the inference snap with OpenCode

Once configured, you can select the inference snap served model within OpenCode.

Use /models in OpenCode and select the newly created model, e.g. gemma4-e4b-q4-k-m Gemma 4 Inference Snap.

Refer to the OpenCode documentation for details on using AI features.