Verda
Verda builds a vertically integrated, sovereign and sustainable full-stack AI cloud, managing data centers, hardware, platform services and AI research. Its goal is to make instant, on-demand GPU compute available at scale while lowering compute's carbon footprint.
At a Glance
- AI startups and software teams
- Frontier AI and foundation-model developers
- Machine-learning research teams
- Enterprises building production inference and AI agents
- +2 more
AI Tools by Verda
(1)
Verda
European AI Cloud GPU Platform
Discussions
No discussions yet
Be the first to start a discussion about Verda
Latest News
Aleph Alpha releases Kolibri, a sovereign open-weight LLM trained on Verda NVIDIA Blackwell clusters
Verda retains Bronze in SemiAnalysis ClusterMAX 3.0
Verda raises $189M/€163M to accelerate its full-stack AI cloud
Verda appoints Mikko Einiö as Chief Financial Officer
Products & Services
On-demand, spot and reserved NVIDIA GPU instances, including GB300, B300, B200, H200, H100, A100, RTX PRO 6000, L40S and other models.
Self-service multi-node GPU clusters with InfiniBand interconnect and pay-as-you-go or reserved capacity.
Dedicated high-clock-speed CPU instances, including standard and confidential-computing variants.
Auto-scaling GPU containers for inference and batch jobs; introduced during the first half of 2025.
Market Position
Verda positions itself as a European full-stack AI cloud and NVIDIA Preferred Partner that owns the data-center, hardware and software stack, rather than only brokering GPU capacity. It competes with hyperscalers and GPU clouds such as AWS, Google Cloud, Microsoft Azure, CoreWeave, Lambda, RunPod and Nebius by emphasizing instant self-service access, renewable energy, European sovereignty, predictable cost and AI-research support.
Leadership
Founders
Ruben Bryon
Seasoned entrepreneur with experience in cloud software and GPU virtualization; founder and CEO who launched Verda's self-service GPU cloud from his Helsinki garage.
Miłosz
Joined Ruben Bryon in the founding days to build the self-service GPU cloud; Verda's official company history identifies him by first name.
Tamir
Joined Ruben Bryon in the founding days to build the self-service GPU cloud; Verda's official company history identifies him by first name.
Executive Team
Ruben Bryon
CEO / Founder
Seasoned entrepreneur with extensive cloud software and GPU virtualization experience.
Arturs Polis
CTO
AI research and hands-on engineering leader building cloud infrastructure from hardware upward.
Founding Story
In 2020, Ruben Bryon started a self-service GPU cloud from his garage in Helsinki after seeing the high costs, complexity and poor customer experience of hyperscaler offerings. Miłosz and Tamir joined him; the business became cashflow-positive within weeks and was initially operated as DataCrunch before rebranding to Verda in November 2025.
Business Model
Revenue Model
Usage-based cloud infrastructure: customers pay hourly for on-demand or spot GPU/CPU instances, multi-node clusters and serverless containers, with reserved capacity discounts and storage billed per GiB-month; enterprise customers can obtain custom quotes.
Pricing Tiers
Pay-as-you-go GPU/CPU and cluster capacity with no commitment.
Discounted interruptible capacity.
Reserved GPU capacity; longer or larger cluster reservations can receive custom quotes.
NVMe, shared filesystem and container registry storage.
Target Markets
- AI startups and software teams
- Frontier AI and foundation-model developers
- Machine-learning research teams
- Enterprises building production inference and AI agents
- Regulated and sovereignty-sensitive organizations
- Media, image, video and interactive-AI companies
- AI model training and foundation-model development
- Large-scale and real-time inference
- Rapid prototyping and experimentation
- Generative video, image and media generation
- AI agents and production ML workloads
- Confidential inference and fine-tuning
- 1X
- Nokia
- ExpressVPN
- Freepik