> ## Documentation Index
> Fetch the complete documentation index at: https://docs.valuez.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Credit Usage
> Understand credit balance, charges, and usage trends
The **Credit Usage** page explains how workspace credits are consumed over time.
Open **Settings & Billing → Credit Usage**. Use it when checking a balance change, planning campaign capacity, or comparing consumption between product areas. The top shows **Available credits | Account credits used | Total account credits** with **Add Credits**, **View usage logs**, and **Download CSV**.
## Filter usage
Filter by date (up to 366 days, default last 30 days) and usage category. Available categories include **All usage, Phone calls, Speech to text, Text to speech, AI responses, Voice agents, Phone numbers, Other**. The summary line shows total credits used, selected tab, date range, ledger entries, and credits returned.
Choose **Auto**, **Daily**, **Weekly**, or **Monthly** grouping for the trend. A wider date range is more useful for budgeting, while a shorter range is better for investigating a recent change. Use **Refresh credit usage** after recent calls; the latest entry may take a short time to appear.
## Read the page
| Section | Purpose |
| - | - |
| Usage breakdown | Compare credits consumed by category with Service, Quantity, % of usage, and Credits used. Quantity shows minutes, characters, tokens, or entry counts |
| Usage trend | See how consumption changes over time with Credits used legend |
| Usage history | Review each charge with Activity, Usage details, Date (DD MMM YYYY, hh:mm A), Session identifier, and Credit change (charges in red, returns in green). Mobile view shows the same as cards |
Activity labels map internal codes to readable text: voice call, outbound / inbound phone call, speech recognition, AI voice generation, AI response, voice test, voice / text / AI response APIs, website voice agent, and virtual phone number rental.
Indicative rates shown in the product include outbound call per minute, inbound call per minute, speech-to-text per minute listened, text-to-speech per 1,000 characters, AI response per 1,000 tokens, and virtual number rental per month (for example 700 credits). Quantity shows the measured usage for the entry, while rate shows how that usage was converted into a credit charge, such as credits per minute, per 1,000 characters, per 1,000 tokens, or per month.
## Exact credit rates
| Activity | Cost | Measured per |
| - | - | - |
| Outbound call | 1.0 credit | 60 seconds |
| Inbound call | 0.5 credits | 60 seconds |
| IVR call segment | 1.0 credit | 60 seconds |
| Speech-to-text (calls and playground) | 0.6 credits | 60 seconds listened |
| Text-to-speech (calls and playground) | 5.0 credits | 1,000 characters (200 characters per credit) |
| AI response (chat, voice, widget, API, tests) | 1.5 credits | 1,000 tokens input + output |
| Virtual number rental | 700.0 credits | Month |
Playground and API variants bill at the same rates as their call equivalents. A single voice call therefore produces several ledger lines - calling minutes, listening minutes, generated characters, and response tokens are metered independently.
Use the session identifier to compare a charge with the related call or activity. Some costs may appear as separate entries because speech, AI response, calling, and number rental are measured independently.
CSV export uses columns **Activity, Source, Usage details, Rate, Date, Session, Credits used, Credits returned** with filename `valuez-credit-usage-{category}-{from}-to-{to}.csv`. Table paging supports **Rows per page 10 / 20 / 25 / 50**.
Compare both values before assuming that a single entry is incorrect.
## Investigate unexpected usage
1. Select the smallest date range that contains the change.
2. Filter to the category with the largest increase.
3. Open individual history entries and note their session identifiers.
4. Compare them with call logs, campaign schedules, or phone-number rental activity.
5. Check whether a recent launch or settings change explains the increase.
## Control future usage
* Test campaigns with a small contact group before a full launch.
* Use appropriate daily and simultaneous call limits.
* Pause campaigns that are producing repeated failures.
* Review unusually long calls and repeated retry attempts.
* Monitor virtual-number rental dates and inactive assignments.
Set a regular review schedule during high-volume campaigns so the team can respond before the workspace runs out of credits.
## Budget a campaign before launch
Estimate credits as contacts × expected connection rate × average minutes × per-minute call rate, plus TTS/STT and AI response overhead per call, plus number rental for the campaign window, plus a 20% reserve for retries and retests. Compare the estimate with available balance and the plan's monthly credits. If the estimate exceeds balance, shrink the first wave rather than launching full-list and stalling mid-campaign - a paused campaign with half-called contacts costs more in confusion than a small clean pilot.
Refresh the page after a recent call if the latest entry has not appeared yet.
Review plan limits, add credits, and check payment history.
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.