toolspool
Verda logo

Verda

verifiedPaidAPItakomo.ai

On-demand NVIDIA GPU cloud offering pay-as-you-go instances, self-service clusters, and serverless containers for AI workloads.

What it does

The site now presents itself as Verda, a European full-stack AI cloud offering NVIDIA GPU instances ranging from A100s to the latest GB300 NVL72, alongside self-service InfiniBand-connected clusters and auto-scaling serverless GPU containers. It also provides high-speed block and shared storage, an OCI-compliant container registry, and runs an in-house AI Lab that collaborates with projects like vLLM and SGLang. Pricing is transparent, pay-as-you-go or spot, by GPU type.

Core features

On-demand NVIDIA GPU instances from A100 to GB300 NVL72
Self-service instant clusters with InfiniBand interconnect
Auto-scaling serverless GPU containers for inference/batch jobs
High-speed NVMe block storage and POSIX shared filesystem
SOC 2 Type II and GDPR-compliant EU-based infrastructure
In-house AI Lab research collaboration and confidential computing option

Best for

ML teams training or fine-tuning foundation models
Startups needing rapid GPU access for prototyping
Companies running scalable GPU-backed inference
Researchers needing hardware-attested confidential compute

Pricing

1x A100 SXM4 80GB
$1.79/hr
1x H100 SXM5 80GB
$3.25/hr
1x B200 SXM6 180GB
$6.11/hr
1x GB300 SXM6 288GB
$8.62/hr

Reviews

Big-picture takes: what it's for and whether it delivers. High-engagement YouTube videos — not sponsored.

Tutorials

Step-by-step: exactly how to get things done with it.