Oblivus

Provides on-demand cloud GPU and CPU compute, storage, and networking for AI training and inference, VFX rendering, cloud gaming, and research workloads, addressing cost and access constraints for high-performance computing. Serves developers, data teams, and IT teams in businesses needing scalable compute from one to thousands of GPUs. Positioned as up to 80% cheaper than major cloud providers with transparent per-minute billing, no quotas, and API-driven provisioning across 9 data centers.

Key features

• On-demand GPU compute
• Scalable CPU compute
• Cloud storage included
• Cloud networking infrastructure
• 9 data centers
• 8 GPU SKUs including RTX 5090
• NVIDIA A6000 instances
• NVIDIA A100 80GB instances
• NVIDIA H100 instances
• Lightning-fast API with 30+ endpoints
• Per-minute billing
• On-demand and reservation pricing
• No quotas or verification
• Flexible instance deploy and rescale
• OblivusAI OS with ML libraries
• VFX cloud rendering capacity
• 99.99% uptime