OpenCode Integration

OpenCode is an AI-powered CLI coding agent. It uses a JSON configuration file to define providers and models.

Configuration File

OpenCode reads from opencode.json in your project root or home directory.

Configuring Providers

All Providers in One Config

{
  "providers": {
    "anthropic": {
      "apiKey": "${ANTHROPIC_API_KEY}",
      "models": {
        "claude-sonnet-4-20250514": {
          "maxTokens": 8192
        }
      }
    },
    "openai": {
      "apiKey": "${OPENAI_API_KEY}",
      "models": {
        "gpt-4o": {
          "maxTokens": 4096
        }
      }
    },
    "google": {
      "apiKey": "${GEMINI_API_KEY}",
      "models": {
        "gemini-2.0-flash": {
          "maxTokens": 8192
        }
      }
    },
    "deepseek": {
      "apiKey": "${DEEPSEEK_API_KEY}",
      "models": {
        "deepseek-coder": {
          "maxTokens": 4096
        }
      }
    },
    "groq": {
      "apiKey": "${GROQ_API_KEY}",
      "models": {
        "llama-3.3-70b-versatile": {
          "maxTokens": 8192
        }
      }
    }
  }
}

Single Provider Examples

Anthropic Claude:

{
  "providers": {
    "anthropic": {
      "apiKey": "${ANTHROPIC_API_KEY}",
      "models": {
        "claude-sonnet-4-20250514": { "maxTokens": 8192 },
        "claude-3-haiku-20240307": { "maxTokens": 4096 }
      }
    }
  }
}

OpenAI GPT:

{
  "providers": {
    "openai": {
      "apiKey": "${OPENAI_API_KEY}",
      "models": {
        "gpt-4o": { "maxTokens": 4096 },
        "gpt-4o-mini": { "maxTokens": 4096 }
      }
    }
  }
}

Google Gemini:

{
  "providers": {
    "google": {
      "apiKey": "${GEMINI_API_KEY}",
      "models": {
        "gemini-2.5-pro-preview-0520": { "maxTokens": 8192 },
        "gemini-2.0-flash": { "maxTokens": 8192 }
      }
    }
  }
}

DeepSeek:

{
  "providers": {
    "deepseek": {
      "apiKey": "${DEEPSEEK_API_KEY}",
      "models": {
        "deepseek-chat": { "maxTokens": 4096 },
        "deepseek-coder": { "maxTokens": 4096 }
      }
    }
  }
}

Groq:

{
  "providers": {
    "groq": {
      "apiKey": "${GROQ_API_KEY}",
      "models": {
        "llama-3.3-70b-versatile": { "maxTokens": 8192 },
        "mixtral-8x7b-32768": { "maxTokens": 4096 }
      }
    }
  }
}

Environment Variables

Set the API keys for whichever providers you use:

export ANTHROPIC_API_KEY="your-anthropic-key"
export OPENAI_API_KEY="your-openai-key"
export GEMINI_API_KEY="your-gemini-key"
export DEEPSEEK_API_KEY="your-deepseek-key"
export GROQ_API_KEY="your-groq-key"

Add to your shell profile to persist across sessions:

echo 'export ANTHROPIC_API_KEY="your-key"' >> ~/.bashrc
source ~/.bashrc

Verification

Confirm your config file is valid JSON:
Terminal window
```
cat opencode.json | python3 -m json.tool
```
Run OpenCode and check available models:
Terminal window
```
opencode
```
Test with a simple prompt to confirm the provider responds correctly.

Troubleshooting

Invalid API key error: Verify the environment variable is set with echo $ANTHROPIC_API_KEY
Model not found: Check the model name matches exactly what the provider expects
Rate limit errors: Switch to a different provider or wait for the limit to reset