Gcore Evolves Everywhere AI into a Full-Lifecycle AI Platform with Slurm, Jupyter, and Token-Based Inference

Enables enterprises to manage development, training, and inference within a single, intuitive platform with native feature integrations 

AMSTERDAM, March 24, 2026 - KubeCon - Gcore, the global infrastructure and software provider for AI, cloud, network, and security solutions, today announced a major evolution of its Everywhere AI solution. With the addition of managed Slurm orchestration, integrated Jupyter development environments, managed NVIDIA Dynamo, and token-based inference usage, Everywhere AI now supports the full lifecycle of AI workloads, from experimentation and training to optimized, consumption-based inference at any scale.

Over the past year, Gcore has steadily expanded Everywhere AI beyond its initial inference capabilities. What began as a Kubernetes-native inference layer has matured into a structured AI execution platform designed to meet the complete operational realities of AI teams. The latest developments enable organizations to manage AI across the full operational lifecycle within a single, Kubernetes-based platform. Rather than deploying separate tools for development, training, and serving, organizations can standardize on Everywhere AI as a managed AI execution layer.

Seva Vayner, Product Director of Edge Cloud and AI at Gcore, comments, “Enterprise AI adoption requires more than raw infrastructure; it demands intelligent orchestration and optimized execution. Everywhere AI has evolved into a unified platform that brings together development workflows, AI applications, and production inference within a Kubernetes-native architecture. With NVIDIA Dynamo integrated as a managed capability and token-based usage models, we provide enterprises with a scalable, performance-optimized foundation for AI across public, private, and hybrid environments.”

Integration JupyterLab to bridge development and production

AI workflows often break down between development and execution environments. Data scientists prototype locally or in isolated notebooks, then hand off workloads to separate infrastructure teams for scaling and deployment.

Integrated Jupyter eliminates that fragmentation. Developers can experiment interactively within the same Everywhere AI platform that supports distributed training and inference. This shortens the path from proof of concept to production deployment and aligns infrastructure more closely with development workflows. For organizations scaling AI initiatives, this native integration reduces friction across teams and supports continuous experimentation.

Native Slurm for integrated training capabilities

Managed Slurm orchestration brings production-grade training capability to Everywhere AI. Distributed AI training requires precise scheduling, efficient GPU allocation, and multi-node coordination — functions traditionally handled by dedicated HPC environments. By integrating Slurm into Everywhere AI, Gcore provides enterprise-ready orchestration without requiring customers to build or manage complex training infrastructure. For companies training at scale, this reduces operational burden and accelerates development speed.

Tokens and Dynamo for flexible and efficient inference deployment

As AI applications move into production and scale, inference costs become a central concern. Many organizations overprovision GPU infrastructure to accommodate peak usage, leading to inefficiencies. Gcore’s introduction of token-based inference usage and managed NVIDIA Dynamo provides a more flexible, cost-effective path to inference deployment. Customers can now consume inference capacity based on actual token usage rather than fixed GPU reservations.

Gcore will showcase its AI solutions at KubeCon, taking place in Amsterdam from 23–26 March 2026. Visit us at booth 1187, Hall 5 to discover more and connect with the Gcore team.

About Gcore

Gcore is a global provider of infrastructure and software solutions for AI, cloud, network, and security, headquartered in Luxembourg. Operating its own sovereign infrastructure across six continents, Gcore delivers reliable, ultra-low latency performance for enterprises and service providers. Its AI-native cloud stack enables organizations to build, train, and scale AI models seamlessly across public, private, and hybrid environments, while integrating AI, compute, networking, and security into a single platform for mission-critical workloads. Learn more at gcore.com.

Gcore press contact

pr@gcore.com