> ## Documentation Index > Fetch the complete documentation index at: https://koreai.mintlify.app/llms.txt > Use this file to discover all available pages before exploring further. # Billing and Usage Monitor and manage credit consumption, tool run usage, and cost breakdowns across your Agent Platform account. *** New accounts receive free credits by default: **75 model credits** and **10,000 tool runs**. Model credits cover model inference and guardrails; tool runs are consumed when invoking tools. Once exhausted, these actions are disabled until you add more credits. To add credits, [contact Support](https://support.kore.ai). ## Accessing Billing and Usage 1. Log in and click **Autonomous Agents** from the module list. 2. Click **Settings** on the top navigation bar. 3. In the left navigation pane, click **Billing** > **Usage**. The Usage page contains five tabs: | Tab | What It Shows | | ------------------ | ----------------------------------------------------- | | **Overview** | High-level summary of runs, credits, and usage trends | | **Tools** | Activity and status for all deployed tools | | **Models** | Computational costs per model and deployment | | **Guardrails** | Guardrail statistics and credit consumption | | **Custom Scripts** | Deployed script statistics and hosting credit usage | ## Credit Usage Summary Widget The **Clock** icon in the top-right corner provides an at-a-glance view of your account's credit status. Click it to see: * **Account ID** — Copy this to share with Support for account debugging. * **Credits** — Total allocation and remaining credits for models, guardrails, and custom scripts. * **Tool Runs** — Total allocation and remaining runs for tool automation flows. Click **Manage billing** in this widget to go directly to the Usage page. The icon color reflects your remaining credit balance: | Icon State | Credit Balance | What It Means | | ---------------------- | ---------------- | -------------------------------------- | | **Green** | > 75% remaining | Healthy balance | | **Yellow/Orange** | 25–75% remaining | Moderate usage | | **Red** | \< 25% remaining | Add credits soon | | **Triangle (warning)** | 0% remaining | All credits exhausted; services paused | ## Alerts and Notifications Agent Platform notifies you proactively as credits and tool runs are consumed. ### Banner Alerts In-app banners appear at the top of every page at these thresholds: | Threshold | Banner Type | Action Prompted | | ---------------------------- | ----------- | ---------------------- | | 20%, 50%, 70%, 80%, 90% used | Removable | Top up or upgrade | | 100% used (credits or runs) | Fixed | Upgrade or top up plan | | Free credits expired | Fixed | Upgrade plan | ### Email Notifications Emails are sent at **50%, 70%, 80%, and 90%** consumption of model credits and tool runs. Each email includes next steps for adding credits via Platform Support. ## Overview Tab The Overview tab summarizes resource consumption across the account: | Metric | Description | | --------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Runs** | Tool runs consumed vs. total allocation (e.g., 45 of 10,000 = 0.45% used) | | **Credits** | Credits used across models, guardrails, and custom scripts vs. total (e.g., 212 of 300 = 70.72% used). Includes a color-coded pie chart; hover to see values per category. | | **Usage Trend** | Chart showing tool runs and model credit consumption over the selected time period | To refresh the **Runs** and **Credits** widgets, hover over their top-right corner and click the **Refresh** icon. ## Tools Tab Displays all tools that have been deployed at least once. Tools in development are excluded. | Field | Description | | ------------------ | ------------------------------------ | | **Total Tools** | Total number of tools in the account | | **Total Runs** | Combined run count across all tools | | **Tool Name** | Name of the tool | | **Runs** | Number of times the tool was invoked | | **Owner Name** | User who created the tool | | **Last Active On** | Date the tool was last used | | **Status** | Deployed, Undeployed, or Deleted | ## Models Tab Displays all open-source and fine-tuned models in the account with their associated computational costs. ### Summary Metrics | Metric | Description | | ----------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Total Models** | Total number of models in the account | | **Total Credits** | Credits used across all models and deployments | | **Fine-Tuning Credits** | Credits consumed for fine-tuning, based on training data size, model complexity, epochs, and hardware | | **Hosting Credits** | Credits for deploying and running models on GPUs. Each GPU instance (replica) counts separately—two A100 replicas consume twice the hosting credits of one. | ### Per-Model Fields | Field | Description | | -------------------- | ---------------------------------- | | **Model Name** | Name of the model | | **Type** | Fine-tuned or Base | | **Credits Used** | Credits consumed by this model | | **Last Deployed On** | Date of the most recent deployment | ### Deployment-Level Drill-Down Click any model row to see individual deployment records: | Field | Description | | ------------------- | ----------------------------------- | | **Name** | Deployment name | | **Type** | Deployment type | | **Credits Used** | Credits consumed by this deployment | | **Last Updated On** | Date of the last deployment update | | **Status** | Deployed, Undeployed, or Deleted | ## Guardrails Tab Displays guardrails used in the account with hardware details and associated costs. Guardrail credits are deducted from your Model credits balance shown in the Overview tab. ### Summary and Per-Guardrail Fields | Field | Description | | -------------------- | ----------------------------------------------- | | **Total Guardrails** | Total number of guardrails in the account | | **Total Credits** | Combined credits used by all guardrails | | **Time** (summary) | Total deployment duration across all guardrails | | **Name** | Guardrail or scanner name | | **Credits** | Credits used to deploy this guardrail | | **Time** (per row) | Duration this guardrail was deployed | | **Last Deployed On** | Date of most recent deployment | | **Status** | Deployed or Undeployed | ### Guardrail Detail Panel Click any row to open a right-side panel showing: * Guardrail name * **Hardware**: CPU type, CPU count, usage duration, and credits used * **Total**: Aggregate credits for the hardware ### Credit Behavior for Guardrails | Condition | Behavior | | ------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------- | | Credits drop below lowest threshold | New guardrail deployments are disabled | | Low credit limit reached during active deployment | Deployment continues; account enters negative credits | | Negative credit limit crossed | Deployment stops with message: *"You've used all your available credits. Please add more credits to your account to continue."* | | Zero credits | All new deployments disabled; active deployments stop when negative limit is reached; negative credits adjusted in the next billing cycle | ### Email Notifications Guardrail billing emails notify administrators about credit usage, negative threshold breaches, and potential deployment interruptions. ## Custom Scripts Tab Displays all custom scripts in the account with deployment credits and hosting details. Scripts are charged per deployment based on the hardware profile used. ### Summary and Per-Script Fields | Field | Description | | ---------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- | | **Total Scripts** | Total number of scripts in the account | | **Hosting Credits** | Total credits used by all scripts | | **Script Name** | Name of the script | | **Language (Version)** | Script language and version | | **Credits Used** | Credits consumed for hardware and resources. See [pricing table](/agent-platform/administration/security-and-control#step-3-resource-allocation). | | **Last Active On** | Most recent date the script was active | | **Status** | Deployment status. [Learn more](/agent-platform/administration/security-and-control#script-statuses). | ### Script Detail Panel Click any row to open a right-side panel showing: * **Hosting Infrastructure**: VCPUs and memory configuration * Hosting time in hours and minutes * Credits consumed per component. See [pricing table](/agent-platform/administration/security-and-control#step-3-resource-allocation). * **Total**: Aggregate credits across all components ### Credit Behavior for Custom Scripts | Condition | Behavior | | ------------------------------------------------- | --------------------------------------------------------------------------------------------------------------- | | Insufficient credits | New deployments disabled in the script wizard | | Low credit limit reached during active deployment | Deployment continues; account enters negative credits | | Negative credit limit crossed | Deployment stops with a failure message | | Zero credits | Projects import as drafts only; deployment actions disabled; negative credit violations stop active deployments | ### Email Notifications Custom script billing emails notify administrators about credit usage, negative threshold breaches, and deployment or undeployment events. Deployment emails also include API endpoint access details. ## Best Practices * Review the **Overview** tab regularly to track trends before credits run low. * Use date filters to compare usage across different time periods. * Check the **Tools**, **Models**, **Guardrails**, and **Custom Scripts** tabs to identify cost-saving opportunities. * Before deploying custom scripts, review the [hardware profile and credits table](/agent-platform/administration/security-and-control#step-3-resource-allocation). * Monitor guardrail deployment duration—hardware costs accumulate over time.