Hyderabad, June 2026: Enterprise AI software leader Pegasystems (Pega) has announced a major advancement in enterprise AI with the launch of Pega Infinity 26, introducing a new approach that eliminates per-token pricing for AI-powered agentic workflows. Revealed at the annual PegaWorld 2026 event, the company’s Pega Predictable AI™architecture aims to help organizations scale AI adoption while maintaining predictable costs and reliable outcomes.
As enterprises increasingly deploy AI agents across customer service, financial services, healthcare, and operational processes, concerns over rising token-based billing and inconsistent AI responses have become significant barriers to large-scale implementation. Pega’s latest innovation addresses these challenges by shifting complex AI reasoning from runtime to design time.
A New Model for Enterprise AI
Traditional AI systems rely heavily on Large Language Models (LLMs) during runtime, generating costs based on token consumption. As workflows become more complex, organizations face escalating expenses and inconsistent outputs due to repeated AI reasoning.
Pega Infinity 26 introduces a different approach. Through Pega Blueprint AI™ and the newly launched Pega Infinity Studio™, AI performs deep reasoning during the workflow design phase. Once deployed, AI agents operate using lightweight semantic models that interpret user intent and execute pre-approved workflows efficiently.
This architecture enables organizations to automate mission-critical business processes such as:
- Customer service requests
- Loan approvals
- Insurance claims processing
- Patient experience management
- Compliance workflows
- Business operations automation
By reducing runtime reasoning requirements, enterprises can significantly lower AI operating costs while maintaining governance and consistency.
Addressing the Growing AI Cost Challenge
The announcement comes at a time when enterprises are reassessing AI spending. Many AI providers have transitioned from fixed subscription models to token-based pricing structures, creating uncertainty around long-term operational costs.
According to Pega, the new Predictable AI architecture eliminates what it describes as the “AI Token Tax,” allowing organizations to focus on business outcomes rather than token consumption.
Instead of charging customers based on the number of AI tokens used, Pega will introduce an outcomes-based pricing model. Organizations will pay a flat fee per completed business case, regardless of how much AI processing occurs behind the scenes.
This model aligns AI investment directly with measurable business value and process completion.
Predictable Outcomes for Regulated Industries
One of the key advantages highlighted by Pega is improved reliability in highly regulated sectors such as banking, insurance, healthcare, and government services.
Rather than allowing AI agents to generate different responses for similar scenarios, the system follows predefined workflows that have been reviewed and approved by organizations. This ensures consistency, compliance, and transparency in decision-making processes.
Where deeper AI capabilities are required—such as document analysis, customer interaction summarization, or data extraction—the system uses targeted AI instructions with clearly defined boundaries to maintain control and predictability.
New AI Token Cost Calculator
To help organizations assess potential savings, Pega also introduced the AI Token Cost Calculator, an interactive tool that compares traditional token-based AI costs with Pega’s outcomes-driven model.
The company claims that enterprises may achieve savings exceeding 20 times compared to conventional AI agent deployments, depending on workflow complexity and operational scale.
The calculator enables businesses to estimate potential reductions in AI spending before deploying large-scale agentic solutions.
Industry Perspective
Speaking at the launch, Alan Trefler, Founder and CEO of Pega, emphasized the importance of delivering business outcomes rather than measuring AI success through token consumption.
He stated that enterprises require AI solutions that provide predictable execution, controlled governance, and sustainable economics, particularly as AI adoption moves from experimentation to production environments.
Industry analysts have also noted the growing demand for accountability in enterprise AI investments as organizations seek measurable returns from their AI initiatives.
Availability
Pega announced that Pega Infinity 26 will be available in the third quarter of 2026. Customers will gain access to the new outcomes-based pricing structure, enabling them to deploy AI-powered workflows without incurring token-based charges.
With enterprises increasingly focused on scaling AI responsibly, Pega’s Predictable AI architecture represents a significant step toward more transparent, cost-efficient, and business-driven AI adoption.