Model Configuration

Configure which language models your agent uses, including provider, model selection, and generation parameters.

LLM Providers

Green Monkey supports multiple LLM providers. Select one in the LLM Connection tab:

Anthropic Claude

ANTHROPIC_API_KEY=sk-ant-...

• Models: Claude 4, Claude 3.5 Sonnet, Claude 3 Haiku
• Best for: Reasoning, long context, safety

OpenAI

OPENAI_API_KEY=sk-...

• Models: GPT-4o, GPT-4, GPT-3.5 Turbo
• Best for: General purpose, wide ecosystem

OpenRouter

OPENROUTER_API_KEY=sk-or-...

• Access 100+ models through one API key
• Best for: Flexibility, trying different models

Ollama (Local)

OLLAMA_URL=http://localhost:11434

• Models: Llama, Mistral, CodeLlama, etc.
• Best for: Privacy, offline use, no API costs
• Requires Ollama  installed locally

Custom Provider

For any OpenAI-compatible API:

CUSTOM_API_URL=https://your-api.com/v1
CUSTOM_API_KEY=your-key

Generation Parameters

Fine-tune how the model generates responses:

Testing Your Connection

After configuring a provider:

1. Click Test Connection in the LLM Connection tab
2. The dashboard sends a test prompt to your provider
3. A successful response confirms the connection is working
4. If it fails, check your API key and network connection

Configuration File

Settings are saved to LLM_CONFIG.md:

# LLM Configuration
 
- **Provider:** anthropic
- **Model:** claude-3-5-sonnet-20241022
- **Temperature:** 0.7
- **Max Tokens:** 4096

Tips

• Start with a test/free-tier key to verify the setup
• Use Ollama for development to avoid API costs
• OpenRouter lets you switch models without changing API keys
• Always use environment variables for API keys in production — never commit them to version control