# Understanding Usage Limits This guide explains how usage works across Zaia plans. It is designed to help users understand what is consumed over time, what defines the structural capacity of a workspace, and how subscription-related limits behave during each billing cycle. This page does **not** focus on pricing or plan comparison. Instead, it explains how to interpret subscription limits and how those limits affect day-to-day platform usage. *** Zaia subscriptions include different types of limits, and not all of them behave in the same way. Some limits are **consumed over time**, such as: * **Zaia LLM Credits** * **Agent Executions** Other limits define the **operational or structural capacity** of the workspace, such as: * **Workflow Concurrency** * **Members** * **Tables** * **Table Rows** * **Knowledge Bases** * **Knowledge Capacity** * **Connections** Understanding the difference between these categories is important to correctly interpret what affects monthly usage and what defines the maximum scale of a workspace. *** ### Monthly Limits and Renewal The following limits are **monthly usage limits** and renew automatically at the beginning of each new billing cycle: * **Agent Executions** * **Zaia LLM Credits** These limits are consumed as the workspace operates throughout the month. Other limits, such as members, tables, knowledge bases, and connections, are not consumed in the same way. Instead, they define the fixed capacity and structure of the workspace according to the active plan. *** ### What Consumes Usage #### Zaia LLM Credits Zaia LLM Credits represent the AI processing capacity available in your workspace when using Zaia’s native LLM providers. Credits are consumed whenever an agent uses an AI model to generate, analyze, transform, or process a response. This means LLM Credits are tied to AI usage, not simply to the existence of workflows, tables, or users. The amount consumed may vary depending on factors such as: * The model being used, such as GPT, Claude, or Gemini * The complexity of the request * The number of reasoning steps, iterations, or tool calls involved If your included Zaia LLM Credits reach zero, your workspace can still continue operating by using **Custom LLM Providers** with your own API key, if configured. #### Agent Executions An Agent Execution is counted each time an agent is triggered to perform an action in the workspace. This may include: * A user message * An automated trigger * An API call * A workflow step that activates an agent Agent Executions define the operational activity of your workspace during each billing cycle. In other words, they measure how often your agents are being activated across the platform. *** ### How Workflows Fit Into Usage #### Workflow Executions Workflows are automated processes that run inside your workspace. All plans allow workflow creation and execution. However, workflows do **not** represent a separate monthly credit bucket by themselves. A workflow may consume other types of usage depending on what happens inside it. For example: * If a workflow triggers an agent, it may consume **Agent Executions** * If a workflow uses AI generation or AI-based processing, it may consume **Zaia LLM Credits** This means workflows are not measured as a standalone billing resource in the same way as credits or executions. Instead, they interact with other usage categories based on their behavior. #### Workflow Concurrency What changes between plans is the number of workflows that can run **simultaneously**. **Workflow Concurrency** defines how many workflows can run in parallel at the same time. This means a workspace may support many workflow runs over the course of a month, while still being limited by how many workflows can be active simultaneously in real time. In practice: * **Workflow Execution** means a workflow is running * **Workflow Concurrency** means how many workflows can run in parallel * A workflow may consume **Agent Executions** and/or **Zaia LLM Credits**, depending on what it does So, workflows are best understood as an automation layer that may trigger usage, while concurrency is the limit that controls how many of them can operate at once. *** ### Structural Workspace Limits Not every subscription limit is a usage counter. Some limits define the maximum scale and structure of your workspace. These limits do not behave like monthly consumption buckets. Instead, they determine how much your workspace can contain or support at a given time. #### Members Member limits define how many users can access and operate the workspace. Members do **not** directly consume Zaia LLM Credits by themselves. Instead, they define the collaboration capacity of the workspace. Workspace roles and permissions may vary, but the active plan determines how many users can be added. #### Tables and Table Rows Table-related limits define how much structured data your workspace can store. * **Tables** define how many datagrids your workspace can have * **Table Rows** define the total number of rows allowed across the entire workspace Table Row limits apply at the **workspace level**, not per individual table. This means your total row capacity can be distributed across tables in different ways depending on your use case. #### Knowledge Bases and Knowledge Capacity Knowledge-related limits define how much reference content your workspace can store for agents to use. * **Knowledge Bases** define how many knowledge bases can exist in the workspace * **Knowledge Capacity** defines the approximate volume of stored content available for retrieval and AI usage These limits are structural and are not consumed like monthly credits. #### Connections Connections define how many external integrations your workspace can configure, such as communication channels, APIs, and third-party tools. Like Members, Tables, and Knowledge Bases, Connections are a structural plan limit rather than a monthly consumption metric. *** ### Workspace and Subscription Rules #### One Plan per Workspace Plans are always linked to a **single workspace**. If your organization manages multiple workspaces, each workspace must have its own active subscription. This means plan limits are not shared across workspaces. #### Managing Your Subscription To manage or upgrade your subscription: 1. Click on your Workspace's name on the top left of the page 2. Go to **Settings → Billing** 3. Click **View Plans** 4. Choose the desired plan 5. Confirm payment and apply the change Only **workspace owners** or **admins** can manage subscription changes. #### Subscription Behavior * **Upgrades** take effect immediately * **Downgrades** take effect on the next billing cycle * Updated limits are applied according to the new plan as soon as the change becomes active * **Zaia LLM Credits** and **Agent Executions** renew automatically every billing cycle * **Custom LLM Providers** can still be used even if included Zaia credits are exhausted *** ### How to Read Plan Limits Correctly When reading a subscription plan, it helps to interpret each type of limit according to its function. #### Monthly usage limits These are consumed during the month and then renewed: * **Zaia LLM Credits** * **Agent Executions** #### Operational concurrency limits These control how much can happen at the same time: * **Workflow Concurrency** #### Structural workspace limits These define how much your workspace can contain or support: * **Members** * **Tables** * **Table Rows** * **Knowledge Bases** * **Knowledge Capacity** * **Connections** This distinction helps avoid confusion between what is **spent**, what is **triggered**, and what is simply **available capacity** under the plan. *** ### Key Takeaways * **Zaia LLM Credits** and **Agent Executions** are monthly usage limits * These limits renew automatically every billing cycle * Workflows do **not** form a separate credit bucket * Workflows may consume **Agent Executions** and/or **Zaia LLM Credits**, depending on what happens inside them * **Workflow Concurrency** controls how many workflows can run at the same time * **Members, Tables, Table Rows, Knowledge Bases, Knowledge Capacity, and Connections** are structural workspace limits * Plans are always tied to a **single workspace** * **Custom LLM Providers** can extend usage even after included Zaia credits are exhausted --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.zaia.app/quick-start/understanding-usage-limits.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.