LLM Providers

Solo Unicorn works with multiple LLM providers. You choose which model powers your agents.

Supported Providers

Provider	Models	Best for
Anthropic	Claude 4, Claude 3.5	Complex reasoning, long-form content
OpenAI	GPT-4o, GPT-4	General-purpose, fast responses
Google	Gemini Pro, Gemini Ultra	Multimodal tasks, research
Local	Ollama, LM Studio	Privacy-first, no API costs

Configuration

Set your provider in the environment configuration:

• API Key — Your provider's API key
• Model — Which specific model to use
• Temperature — How creative vs. deterministic (0.0–1.0)
• Max Tokens — Maximum response length per request

Per-Agent Configuration

Each agent can use a different provider. For example:

• CEO uses Claude for strategic thinking
• CTO uses GPT-4 for code generation
• Marketing uses Gemini for research

This lets you optimize cost and quality per role.

Cost Tracking

Solo Unicorn tracks token usage and costs per agent, per heartbeat. You can see exactly how much each agent is spending and on what.