Skip to main content

AI Configuration in Settings

The AI Configuration tab is where you manage your AI provider API keys and default model selection.

Accessing AI Settings​

Go to Settings → AI Configuration tab.

Provider Tabs​

Click the provider name (OpenAI, Anthropic, Google) to switch between them.

Adding a Key​

  1. Select the provider tab
  2. Paste your API key in the input field
  3. Choose your preferred model from the dropdown
  4. Click Validate — YesBill sends a test request to verify the key works
  5. Click Save

A green checkmark (✅) appears next to a validated provider.

Removing a Key​

  1. Open the provider tab
  2. Click Remove Key or clear the field
  3. Save

Default Model​

The model you select here becomes the default in Ask AI and the Agent. You can always change it per-conversation using the model selector in the chat interface.

Model Capabilities​

Some models support Extended Thinking (reasoning) — these show a 🧠 icon in the model selector. Reasoning models think through problems step by step before responding, which can improve accuracy for complex queries.

Usage Monitoring​

Go to Analytics → AI Usage to see:

  • Tokens used per conversation
  • Cost per message
  • Total monthly AI spend

This helps you estimate your AI provider bill.

tip

For everyday use, GPT-4o Mini (OpenAI) or Gemini 2.0 Flash (Google) are excellent budget-friendly choices. Switch to a more powerful model for complex analysis tasks.

Ollama (Local Models)​

YesBill supports Ollama — run AI models entirely on your own machine with no API key and no usage costs.

To use Ollama you need to:

  1. Install Ollama and pull a model
  2. Start Ollama with the correct host binding
  3. Expose it via a Cloudflare Tunnel so YesBill's backend can reach it
  4. Paste the tunnel URL in Settings → Ollama Base URL → Fetch Models → Save

See the full step-by-step guide: Ollama Setup →