Auto-discovers models from Tama local AI server and provides OpenCode with model configuration.

• Auto-detection: Finds tama running on default ports (11434, 8080)
• Model Discovery: Queries /tama/v1/opencode/models for rich model metadata
• Configuration Enhancement: Adds model metadata (context limits, name, etc.)
• Graceful Fallback: Works even if tama is offline

Add to your opencode.json:

{
  "$schema": "https://opencode.ai/config.json",
  "plugin": ["opencode-tama"]
}

Or install via npm:

npm install opencode-tama

Simply install the plugin - it will auto-detect tama and discover models.

If you want to use a custom tama instance:

{
  "provider": {
    "tama": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Tama (local)",
      "options": {
        "baseURL": "http://localhost:11434/v1"
      }
    }
  }
}

The plugin will still enhance this with auto-discovered models, merging with any manually configured ones.

If your tama instance is gated behind a bearer token (e.g. a public endpoint fronted by a reverse proxy), set the token in one of two ways:

1. TAMA_TOKEN environment variable (highest priority):

   export TAMA_TOKEN=your-token-here

2. apiKey in your opencode.json provider options:

   {
     "provider": {
       "tama": {
         "npm": "@ai-sdk/openai-compatible",
         "options": {
           "baseURL": "https://tama.example.com/v1",
           "apiKey": "your-token-here"
         }
       }
     }
   }

The token is sent as Authorization: Bearer <token> on both model discovery and inference requests. When unset, no auth header is sent (fine for localhost).

1. On opencode startup, the config hook is called
2. Plugin checks for existing tama provider or auto-detects on default ports
3. Queries GET /tama/v1/opencode/models from tama
4. Merges discovered models into opencode's configuration
5. Models appear in /models list automatically

• Tama running with tama serve
• OpenCode with plugin support

MIT