Cloud Temple GPU Instances

Cloud TempleCompute

NVIDIA GPU-accelerated compute instances for AI training and inference workloads that need sovereign-FR residency — Cloud Temple's answer to workloads that would otherwise need US-hosted GPU capacity

FluffyStack tools
Add to Service BuilderAdd to CompareCompare with equivalentsExplore Cloud Temple in TreemapExplore compute in HoneycombSee Cloud Temple regions on the World MapSee compute as a networkScore jurisdiction exposure
DocumentationPricingCloud Temple website

Jurisdictional exposure

Provider HQ
EUParis, France

Subject to GDPR, Data Act, DGA

Region locations
EU2 regions
Sovereign option
Yes — 2 sovereign-flagged regions available
Full scorecard for this service →EU lens detail →Sovereign cloud coverage map →

Attributes

GPU Support
Yes

Sub-services (2)

Training-Class GPUs

Higher-VRAM configurations for model training runs

Inference-Class GPUs

Cost-optimised GPUs for serving trained models

Compliance & Certifications

This service is attested for the following frameworks. Always verify with the provider before relying on a specific compliance posture.

GDPRISO 27001SecNumCloud

Where this runs

2 regions
1 country
2sovereign
Sovereign regions (2)
  • Cloud Temple Paris · ParisSecNumCloud
  • Cloud Temple Marseille · MarseilleSecNumCloud

Tags

gpunvidiaaicomputefrancesovereign-frsecnumcloud

Equivalent services on other platforms

OVHcloud AI TrainingOVHcloud

Managed GPU-backed training jobs for custom ML models, with H100 / H200 / L40s / L4 GPU options, distributed-training orchestration, and Jupyter integration

Generative APIsScaleway

Managed inference for open-source LLMs (Llama, Mistral, DeepSeek) hosted in EU datacentres

Pricing

Pricing model:hourly-gpu