opencode-autodiscover

Auto-discover models from local OpenAI-compatible API endpoints for OpenCode.

Features

• Auto-discovers models from local endpoints (Ollama, LM Studio, LLM proxies)
• Fetches metadata from Models.dev and OpenRouter (context window, capabilities, cost)
• Caches metadata locally (24h TTL)
• Supports multiple endpoints
• API key authentication via config or environment variables
• Model filtering with include/exclude patterns

Installation

opencode plugin opencode-autodiscover

Or install globally:

opencode plugin opencode-autodiscover -g

Configuration

Add provider(s) to your opencode.json:

{
  "provider": {
    "local-ollama": {
      "name": "Ollama",
      "options": {
        "baseURL": "http://localhost:11434/v1"
      }
    },
    "local-lmstudio": {
      "name": "LM Studio",
      "options": {
        "baseURL": "http://localhost:1234/v1"
      }
    }
  }
}

With API Key

{
  "provider": {
    "local-proxy": {
      "name": "My Proxy",
      "options": {
        "baseURL": "http://localhost:8080/v1",
        "apiKey": "sk-your-key"
      }
    }
  }
}

Environment Variables

API keys can be set via environment variables:

export OPENCODE_LOCAL_MY_PROXY_API_KEY=sk-your-key

Model Overrides

Override metadata for specific models:

{
  "provider": {
    "local-ollama": {
      "options": {
        "baseURL": "http://localhost:11434/v1",
        "modelOverrides": {
          "my-custom-model": {
            "contextWindow": 128000,
            "maxOutput": 8192
          }
        }
      }
    }
  }
}

Model Filtering

Control which models are discovered using include and exclude glob patterns inside options:

{
  "provider": {
    "local-ollama": {
      "options": {
        "baseURL": "http://localhost:11434/v1",
        "include": ["qwen/*"],
        "exclude": ["*embedding*", "*test*"]
      }
    }
  }
}

• include — if set, only models matching at least one pattern are discovered
• exclude — models matching any pattern are skipped (applied after include)
• * matches anything (including /)
• Matching is case-insensitive
• Special characters (., [, ], etc.) are treated as literals

Usage

Models are auto-discovered at startup and appear in /models natively.

To refresh models, use the refresh-local-models tool in OpenCode, then restart.

CLI Options

opencode plugin opencode-autodiscover         # Install to project config
opencode plugin opencode-autodiscover -g      # Install to global config
opencode plugin opencode-autodiscover -f      # Force update existing plugin
opencode plugin opencode-autodiscover --pure  # Run without external plugins

How It Works

1. On startup, the plugin reads provider configs with baseURL options
2. Fetches available models from each endpoint (/v1/models)
3. Looks up metadata from Models.dev and OpenRouter (context window, cost, capabilities)
4. Injects discovered models into the provider config
5. Models appear in OpenCode's /models UI

Caching

Metadata is cached locally for 24 hours:

• Location: ~/.cache/opencode-autodiscover/
• Files:
  • openrouter.json — OpenRouter metadata cache
  • modelsdev.json — Models.dev metadata cache

Use the refresh-local-models tool to clear cache and re-fetch.

Development

npm install
npm test
npm run build

License

MIT