As enterprises expand the use of artificial intelligence across departments, managing the costs and risks associated with large language models is becoming a growing operational challenge. Parallel Works is addressing that issue with new governance and budget management capabilities designed to help organizations control AI consumption across both commercial and privately hosted models.
The company announced enhancements to its ACTIVATE AI platform that enable enterprises, government agencies and research organizations to manage, monitor and govern AI usage through a centralized gateway. The new functionality allows organizations to oversee access to multiple large language models while tracking usage, controlling costs and applying governance policies across cloud and on-premises environments.
“Organizations are discovering that the future of AI will be defined as much by governance and economics as by the model itself,” said Matthew Shaxted, chief executive officer of Parallel Works.
The announcement comes as businesses increasingly deploy AI services from multiple providers while also investing in privately hosted models and GPU infrastructure. As AI adoption expands, organizations are facing rising expenses tied to token consumption, inference workloads and model access, often without centralized visibility into how those resources are being used.
Parallel Works said its ACTIVATE AI Gateway is designed to bring the same governance principles commonly applied to compute and storage resources to AI environments. The platform provides a unified API gateway that connects commercial AI services and self-hosted models while allowing organizations to apply consistent management and financial controls.
A key feature of the new release is token budgeting and consumption tracking, which gives organizations real-time visibility into AI usage and associated costs. The platform supports governance at the user, team, department and organizational levels, enabling administrators to monitor activity and allocate budgets accordingly.
The company also introduced chargeback and cost-accounting capabilities intended to help organizations assign AI expenses to specific business units and projects. By linking consumption directly to users and departments, enterprises can gain greater accountability as AI deployments scale.
Parallel Works said the platform supports OpenAI-compatible providers, Anthropic, Azure OpenAI, Amazon Bedrock and privately hosted large language models. The vendor-neutral approach allows organizations to use multiple AI services through a single management layer while reducing dependence on any individual provider.
Beyond AI governance, ACTIVATE integrates compute orchestration, GPU resource management, Kubernetes administration and storage governance into a unified platform. The company believes that combining those capabilities can simplify operations for organizations managing increasingly complex AI and high-performance computing environments.
The governance capabilities are already being used within a large system-integrator environment operated by FutureTech, according to the company. The deployment supports thousands of users and manages token consumption across a variety of AI workloads spanning cloud and on-premises infrastructure.
“Our customers are demanding stronger AI governance capabilities to ensure AI can be deployed securely, responsibly, and at scale,” said Chris Coker, vice president of major accounts for aerospace and defense at FutureTech. “The combination of token budgeting, usage visibility and chargeback, integrated directly into the compute governance environment, gives our clients the controls they need to scale AI responsibly and with confidence.”
As organizations seek access to increasingly advanced AI models, technology leaders are also confronting challenges related to spending oversight, security and operational governance. Without centralized controls, AI usage can quickly become fragmented across teams and platforms, making costs difficult to predict and manage.
“Developers consistently want the state of the art, and in AI, that’s changing day by day,” said Michael McQuade, director of engineering at Parallel Works. “Enterprises want to expand AI access across their teams, but without governance controls, costs and operational risks spiral fast.”
The new governance and token budgeting features are available immediately and are targeted at large enterprises, government and defense agencies, high-performance computing environments and research institutions operating private GPU infrastructure or consuming AI services at scale.
Parallel Works provides hybrid multi-cloud computing management software through its ACTIVATE platform, helping organizations provision, manage and share computing resources across cloud and on-premises environments while maintaining visibility into costs, performance and resource utilization.