← Back to Documentation

Providers

PersistenceAI supports multiple LLM providers. You can configure API keys for any provider you want to use.

Connecting a Provider

• Run the /connect command in the TUI

• Select a provider from the list

• Follow the provider-specific authentication steps

• Enter your API key when prompted

/connect

Supported Providers

PersistenceAI supports all major LLM providers:

• OpenAI - GPT-4, GPT-3.5, and more
• Anthropic - Claude models
• Google - Gemini models
• Ollama - Local models (self-hosted)
• PersistenceAI Zero - Curated list of tested models

• And many more...

Provider Configuration

Provider configurations are stored in your global config file:

• Windows: %USERPROFILE%\.config\pai\pai.json
• Linux/macOS: ~/.config/pai/pai.json

Example Configuration

{
  "providers": {
    "openai": {
      "apiKey": "sk-..."
    },
    "anthropic": {
      "apiKey": "sk-ant-..."
    },
    "ollama": {
      "baseURL": "http://localhost:11434"
    }
  }
}

Ollama (Self-Hosted)

PersistenceAI has excellent support for Ollama, allowing you to run models locally without API costs.

Setup

• Install Ollama from ollama.ai

• Pull the models you want to use:

ollama pull llama2
   ollama pull codellama

• Connect to Ollama in PersistenceAI:

/connect

Select "Ollama" and enter your base URL (default: http://localhost:11434)

Recommended Models

• cogito:3b - Fast, lightweight for simple tasks
• qwen2.5-coder:4b - Good balance of speed and capability
• qwen2.5-coder:7b - More capable for complex tasks
• qwen2.5-coder:13b+ - Maximum capability

PersistenceAI Zero

PersistenceAI Zero is a curated list of models that have been tested and verified by the PersistenceAI team.

• Run /connect and select "PersistenceAI Zero"

• Visit the authentication page

• Sign in and get your API key

• Paste the key in PersistenceAI

Model Selection

You can switch between models using:

• Keyboard: F2 to cycle through recently used models
• Command: /models to see all available models
• Shortcut: M (default: Ctrl+X M)

Provider-Specific Notes

OpenAI

• Requires API key from platform.openai.com

• Supports GPT-4, GPT-3.5, and other models

• Usage-based billing

Anthropic

• Requires API key from console.anthropic.com

• Supports Claude 3 models

• Usage-based billing

Google

• Requires API key from Google AI Studio

• Supports Gemini models

• Usage-based billing

Ollama

• Free and self-hosted

• No API key required

• Runs models locally on your machine

• Best for privacy-sensitive projects

---

For more information, see:

• Configuration - Full configuration options
• Models - Model selection and usage