AI Usage & Cost Statistics
Monitor how your AI features are being used and estimate the costs associated with each AI request.
What You'll See#
The AI Usage panel displays:
| Metric | Description |
|---|---|
| AI Requests | Total number of AI chat questions asked by readers |
| Monthly Limit | Maximum requests allowed on your current plan |
| Daily Trend | Graph showing request volume over time |
| Cost Breakdown | Estimated costs per AI provider and model used |
| Total Cost | Cumulative estimated cost for all AI requests |
| Translation Usage | Number of document translations completed |
How to Open#
- Open any page of your documentation site
- Click the Analytics tab in the floating toolbar at the bottom
- Select the AI Usage section to view your AI metrics
Cost Estimation#
Docsbook estimates costs based on:
- Token counts: Approximated from text length (~4 characters per token)
- Model pricing: Current market rates for popular models (OpenAI, Anthropic, Gemini, etc.)
- Per-request breakdown: Input tokens (question) × input price + output tokens (answer) × output price
Supported pricing models:
| Provider | Models |
|---|---|
| OpenAI | GPT-4, GPT-4o, GPT-4o mini, GPT-3.5 Turbo |
| Anthropic | Claude 3 Opus, Sonnet, Haiku |
| Gemini Pro, Gemini Pro Vision | |
| Others | OpenRouter and custom models (default rates) |
Note: Estimated costs are approximate and may differ from actual provider invoices due to token counting differences, special pricing, and volume discounts.
Monthly Limits#
AI usage is metered in money, not in requests. Each paid plan includes a monthly AI budget equal to its price:
| Plan | Monthly AI budget |
|---|---|
| Free | $0.15 |
| Pro | $59.00 |
| Business | $159.00 |
| Growth | $349.00 |
| Scale | $899.00 |
Every call is charged at the model provider's real price plus a 150% markup, so a cheaper model stretches the same budget further.
When a paid plan exceeds its monthly budget, additional usage is billed as overage up to a monthly cap you set in the Limits tab of workspace settings (default $20/month) — see Pricing & Plans for how the cap works. The Free plan has no overage and stops at its budget.
Translation Usage#
Translation requests use a separate monthly quota:
| Plan | Translations/Month |
|---|---|
| Free | 0 translations |
| PRO | 50 translations |
| PRO+ | 500 translations |
Learn more: AI Translations
Cost Breakdown by Model#
The cost breakdown table shows your spending by AI provider and model:
- Provider: Which AI service (OpenAI, Anthropic, Gemini, etc.)
- Model: The specific model used (e.g.,
gpt-4o-mini) - Count: Number of requests using this model
- Estimated Cost: Total cost for requests using this model
The models are sorted by total cost, so you can quickly see which models are driving your AI expenses.
Custom API Keys#
If you've set your own API keys for OpenAI, Anthropic, or Gemini, the costs are:
- Calculated by the provider (not estimated by Docsbook)
- Visible in your provider's billing dashboard (OpenAI Platform, Anthropic Console, Google Cloud)
- Not included in Docsbook usage limits (you manage costs separately)
Learn more: Using Your Own API Keys
Time Ranges#
AI usage metrics are available for:
| Range | Plan required |
|---|---|
| Last 24 hours | Free |
| Last 7 days | Pro |
| Last 30 days | Growth / Business / Scale |
Start using AI chat today. Create a documentation site →