Verda

Verda (formerly DataCrunch) is a Europe-built AI cloud platform that provides self-service GPU and CPU compute, serverless containers and managed storage. The company says it supports the full AI lifecycle, from rapid prototyping to foundation model training and scalable inference.

What It Is

Verda is a cloud infrastructure provider for AI workloads. Users sign up through a console to provision compute, storage and container services, and can also control GPU resources programmatically through an API. The platform is vertically integrated, spanning data centers, GPU hardware, orchestration and managed services.

Compute, Serverless and Storage

Compute options include instant clusters (self-service GPU clusters with InfiniBand interconnect), GPU instances with NVIDIA hardware, and CPU instances with dedicated cores. Serverless containers offer auto-scaling GPU containers for inference and batch jobs. Storage covers NVMe block storage, a POSIX-compliant shared filesystem and an OCI-compliant container registry. Listed GPUs include GB300 NVL72, B300, B200, RTX PRO 6000, H200, H100 and A100.

AI Lab and Confidential Computing

Verda runs an in-house AI Lab that works on customer co-research, large-scale training, inference optimization, AI compilers and kernel engineering. It also lists a confidential computing offering for hardware-attested inference and fine-tuning. Verda says it sponsors open-source projects such as SGLang and vLLM with compute access.

Security and Compliance

Verda states it is SOC 2 Type II and ISO 27001/27017/27018/27701 certified, GDPR compliant and powered by renewable energy, and describes itself as an NVIDIA Preferred Partner.