Ask AI

Ask AI is a persistent chat interface where you can have conversations about your bills, services, and spending patterns in plain English.

Opening Ask AI

1. Click Ask AI in the sidebar
2. Type your question in the chat input
3. Press Enter or click Send

Asking Questions

Ask AI understands natural language. Simply type as you would to a person.

Bill Queries

• "How much did I spend on all services in February?"
• "What's my total milk bill for this year?"
• "Did I pay the internet bill in December?"

Service Tracking

• "How many days was the tiffin delivered in January?"
• "Did the maid miss any days last week?"
• "Which services are currently active?"

Spending Analysis

• "What's my most expensive service?"
• "How has my monthly spend changed over the last 6 months?"
• "Break down my household expenses by category"

General Help

• "How do I add a new service?"
• "What's the difference between Visit-based and Home Delivery?"

---

Chat Options & Conversations

Each conversation is saved automatically. The left panel shows your conversation history — click any past conversation to resume it.

Right-click or hover over a conversation in the history list to access options: Rename or Delete a conversation.

---

Per-Message Features

Each AI response has a set of action buttons below it:

Button	What It Does
📊 Analytics	Shows token usage and cost for that specific message
👍 / 👎 Feedback	Rate the response quality (helps improve the AI)
📋 Copy	Copy the full response text to your clipboard

---

Model Selection

Use the model selector above the chat input to choose which AI model handles your query. Available models are based on which API keys you've configured:

• OpenAI — GPT-4o, GPT-4o Mini, GPT-4.1
• Anthropic — Claude 3.7 Sonnet, Claude 3.5 Haiku
• Google — Gemini 2.0 Flash, Gemini 1.5 Pro

Switch models freely between messages — each conversation remembers which model was used for each response.

---

Reasoning Effort

For models that support extended thinking (Claude 3.7 Sonnet, o3-mini), a Reasoning selector appears next to the model picker:

Level	Speed	Depth	Use When
Low	Fastest	Surface	Quick factual lookups
Medium	Balanced	Standard	Most everyday questions
High	Slowest	Deep	Complex multi-month analysis

---

Token Usage & Thinking Analytics

Click the 📊 icon on any message to open the Analytics panel showing:

• Model used and provider
• Prompt tokens and completion tokens
• Estimated cost for that query

When using a reasoning model (High effort), the analytics panel also shows the thinking token count — the internal reasoning steps the model performed before responding. Thinking tokens consume more budget but produce more accurate, nuanced answers.

tip

Ask AI has access to all your YesBill data — services, tracking history, bills, and payments. The more data you've entered, the better the answers.