Set Up Your LLM

Kakaw supports five LLM providers. You supply the API key — it goes directly from your browser to the provider. Kakaw never sees or stores it.

Anthropic (Claude)

1. Get an API key at console.anthropic.com
2. In Kakaw, go to Settings → LLM Connections
3. Select Anthropic, paste your key, and click Save

Recommended model: claude-sonnet-4-6 — fast, accurate for SQL generation.

OpenAI (ChatGPT)

1. Get an API key at platform.openai.com
2. In Kakaw, select OpenAI, paste your key, and click Save

Recommended model: gpt-4o

Gemini (Google)

1. Get an API key at aistudio.google.com
2. In Kakaw, select Gemini, paste your key, and click Save

Recommended model: gemini-1.5-pro

Ollama (local)

1. Install Ollama from ollama.com and pull a model: ollama pull llama3
2. Ensure Ollama is running: ollama serve
3. In Kakaw, select Ollama and enter your local endpoint (default: http://localhost:11434)

No API key required for Ollama.

LM Studio (local)

1. Download LM Studio from lmstudio.ai and load a model
2. Start the local server in LM Studio (default port: 1234)
3. In Kakaw, select LM Studio and enter your endpoint (default: http://localhost:1234/v1)

No API key required for LM Studio.

Switching Providers

You can change your LLM provider at any time in Settings → LLM Connections. Your conversation history is preserved regardless of which provider you switch to.