Scaleway: what to verify before choosing

Scaleway offers GPU Instances that provide dedicated GPUs within virtual compute instances and Managed Inference/Generative APIs using dedicated GPU infrastructure; its cloud also has separate Kubernetes and serverless products. A GPU Instance gives direct SSH/container control; Managed Inference provides a managed model endpoint. GPU Instances can be attached to orchestration separately, with corresponding node and operations responsibilities.

Updated

Computraq is preparing for launch. These guides explain readiness and the proposed framework; they do not establish a live service, integration or partnership.

Delivery modes and operating boundary

Scaleway offers GPU Instances that provide dedicated GPUs within virtual compute instances and Managed Inference/Generative APIs using dedicated GPU infrastructure; its cloud also has separate Kubernetes and serverless products.

A GPU Instance gives direct SSH/container control; Managed Inference provides a managed model endpoint. GPU Instances can be attached to orchestration separately, with corresponding node and operations responsibilities.

Computraq is prelaunch. Naming this provider here does not mean that Computraq hosts workloads on it, has tested its integration, or has a commercial relationship with it.

Buyer and procurement preparation

Use these questions to prepare a comparable, workload-specific evaluation. Product names describe the provider’s published offering; they do not establish availability for a particular account.

  • Select direct GPU Instances for custom environments or Managed Inference for a managed model/API; check availability zones and current stock for the chosen accelerator.
  • Confirm data-residency and private-network requirements against the specific product, as inference and compute services can have different controls.
  • Estimate compute, storage, public IP, network and managed-service costs from current zone-specific terms.

Developer and operator evidence

Before selecting capacity, create a small acceptance plan that records the model and revision, input shape, runtime, GPU configuration, concurrency, target latency or job completion, and recovery expectations. Provider documentation is evidence of product design, not proof that this workload has been tested.

  • For instances, choose a supported GPU OS/container image, verify driver/runtime, SSH exposure and volume sizing.
  • For Managed Inference, test the OpenAI-compatible interface, model catalog or custom model, throughput, networking and privacy assumptions.
  • If running GPU nodes in Kubernetes, include instance provisioning and node setup time in workload capacity planning.

Sources and verification