Licensing for conversational agents

Hybrid licensing model for conversational agents, covering consumption factors, available plans, and deployment planning scenarios.

Conversational agents use a hybrid user- and consumption-based licensing model. This page explains how licensing works, what factors affect consumption, and provides scenarios to help you plan your deployment.

Pricing models overview

UiPath offers two licensing plans for conversational agents. Both use a consumption-based model where usage is measured in exchanges.

What counts as an exchange

An exchange includes:

• The user's prompt.
• The agent's response(s).
• Any tool calls executed in between.

Exchanges do not draw from the monthly limit when the user's prompt contains eight characters or fewer and no tool calls are executed.

AI and Agentic usage pool

All Enterprise licensed users receive a monthly AI and Agentic usage pool, shared across UiPath AI and Agentic products and usable for conversational agents. The pool replaces the previous per-user entitlement of 50 free exchanges per month.

When you use the AI and Agentic usage pool for conversational agents, usage is drawn from the pool per exchange, based on the cost of that exchange. The pool capacity varies by user license, as shown in the following tables.

For details on how the monthly limit, top-ups, and paid-unit fallback work, refer to AI and Agentic usage pool.

Note:

The AI and Agentic usage pool is available on the Unified Pricing and Flex licensing plans, and is not available on Community. Feature availability also depends on the cloud platform that you use; for details, refer to the Feature availability page.

Flex licensing

Flex licensing uses Agent Units for consumption. When the monthly limit and any top-up are spent, conversational agents continue by drawing from Agent Units, at 1 Agent Unit per exchange.

User license	AI and Agentic usage pool
Express User (formerly known as Automation Express and Autopilot Express)	No monthly limit; 50 free exchanges per user per year, then 1 Agent Unit per exchange
Cloud Basic User	Basic Usage Pool
Attended - Named User	Basic x7.5
Citizen Developer	Basic x7.5
Automation Developer	Basic x20

Unified Pricing

Unified Pricing uses Platform Units for consumption. When the monthly limit and any top-up are spent, conversational agents continue by drawing from Platform Units at the Conversational Agent Rate.

User license	AI and Agentic usage pool
Basic User	Basic Usage Pool
Plus User	Basic x7.5
Pro User	Basic x20
App Tester	Basic x7.5
App Tester Developer	Basic x20

The rate at which the fallback draws varies by the LLM model tier configured for the agent:

Model tier	Consumption rate
Standard models	0.2 Platform Units (referred to as Conversational Agent Rate)
Basic models	0.16 Platform Units (Conversational Agent Rate × 0.8)

Autopilot for Everyone

Autopilot for Everyone also draws from the monthly AI and Agentic usage pool. Once the monthly limit and any top-up are spent, it is charged per action at the Conversational Agent Rate (0.2 Platform Units per action on Unified Pricing). For details, refer to Autopilot for Everyone licensing.

External users (no license)

For anonymous access authentication:

• No user license required.
• Requires an external application token (created in Admin portal).
• Each exchange consumes from the tenant's unit pool starting with the first exchange.
• Deploy via iFrame embedding with an auth server.

Design time entitlement

Development and debugging of agents draws from the same monthly AI and Agentic usage pool as published agent exchanges. Design time usage is no longer measured as a separate daily LLM call allotment, because the monthly limit is shared across all AI and Agentic products and activities.

Exchange consumption by user license

The following table shows how consumption works within and beyond the monthly limit:

User license	Within the monthly limit	After the monthly limit is spent
Express User	50 free exchanges per user per year	Consumes from the tenant pool
All other licensed users	Draws from the monthly limit, sized by license	Any administrator-granted top-up, then the tenant pool
External user (no license)	Not applicable	Consumes from the tenant pool from the first exchange

Tool consumption

While some tool costs are included in the base exchange, certain tools have additional consumption requirements. Tools are billed per activity executed, not per connector configured.

Tool	Flex (Agent / Misc Units)	Unified (Platform Units)
Analyze files	1 AU per file	0.2 PU per file
IXP tool	1 AU per page	0.2 PU per page
MCP tool	Included	Included
Context Grounding (semantic search)	Included	Included
Context Grounding (DeepRAG)	1 AU per ~20 pages	0.2 PU per ~20 pages
Integration Service activity	Included	Included
API workflow	5 IS API calls per activity executed	0.02 PU per activity executed
Cross-platform RPA	1 Robot Unit	0.2 PU
Windows RPA (Unattended)	Requires Unattended Robot	Requires Unattended Robot
Autonomous Agent (nested)	Nested agent's consumption	Nested agent's consumption

Scenarios

The following scenarios illustrate how licensing works in practice.

Scenario 1: Simple Q&A assistant

Agent configuration

• Tools: Context Grounding semantic search only
• No additional tools

Consumption calculation

Because this example uses only Context Grounding semantic search, there is no additional tool cost per exchange.

Note:

Monthly limits are not pooled across users. Each user has their own monthly limit, consumable only by them. If User A uses a fifth of their limit while User B exhausts theirs, User A still has four fifths remaining, and User B's further exchanges draw from a top-up if an administrator granted one, and then from the tenant pool.

Example: Usage with different license types:

• User A (Basic User): around 20 exchanges this month, so roughly 40% of the monthly limit used
• User B (Basic User): around 100 exchanges this month, so the monthly limit is exhausted after roughly 50 and the remainder draws from the tenant pool
• User C (Pro User): around 500 exchanges this month, so roughly half of the monthly limit used
• Total tenant pool consumption: only User B's exchanges beyond their monthly limit

For customers who want a pure consumption-based model without maintaining Automation Cloud user accounts, deploy your conversational agent via iFrame embedding using anonymous access. This begins consuming units immediately with no free tier.

Scenario 2: Self-service assistant with multiple tools

Agent configuration

• Context Grounding semantic search
• Analyze files tool
• Integration Service connector
• API workflow with two connector activities executed

Consumption calculation

Additional tool costs

Interaction type	Consumption (Flex)	Consumption (Unified)
Semantic search	Included	Included
File analysis	1 AU per file	0.2 PU per file
IS connector	Included	Included
API workflow (2 activities executed)	10 IS API calls	0.04 PU

Example: Monthly usage with different license types and tools:

User	License	Exchanges	Flex consumption	Unified consumption
A	Express	30 (20 Q&A, 10 file analysis)	30 free exchanges, 10 AU for file analysis	30 free exchanges, 2 PU for file analysis
B	Basic	75 (50 Q&A, 15 file analysis, 10 API workflows)	Exchanges draw from the monthly limit, then 1 AU each; 15 AU tools + 100 IS API calls	Exchanges draw from the monthly limit, then 0.2 PU each; 3.4 PU tools
C	Plus	200 (mixed tools)	Exchanges draw from the monthly limit, tool consumption applies	Exchanges draw from the monthly limit, tool consumption applies

Bring your own model (Preview)

Conversational agents support LLM Configurations so you can use your own LLM subscriptions. This feature is in preview. During the preview period, there is no charge per exchange, however additional consumption based on tool usage still applies.

Managing consumption

Admin controls

Administrators can manage consumption at the tenant level:

• Monitor usage through the Admin portal.
• Set up quotas for consumption thresholds.
• Review usage by agent.

Next steps

• Deployment: Deploy your agent
• Limitations and FAQ: Current limitations and troubleshooting