AI Cloud Pricing
With vertical integration and continuous optimization of operational costs, we provide predictable, competitive pricing that maximizes value for your AI workloads.
Commitment discounts
Save up to 35% on on-demand rates when you reserve large-scale clusters for a multi-month period.
Flexible pricing strategy
Benefit from combining discounts on commitment usage and scalability-friendly, on-demand rates.
Convenient payment methods
Choose the payment method that best suits your needs: credit cards or bank transfers.
Pricing that scales from a single GPU to enterprise infrastructure.
Start with flexible access for development, move into reserved capacity for production, or deploy Istidlal as your enterprise inference platform.
Developers
For individual engineers and builders who need fast access to production-grade GPUs.
Use full or fractional GPUs on demand, launch containers or inference workloads, and reserve GPU-backed VMs when you need more control.
- Full or sliced GPU access
- On-demand usage for experiments and apps
- Reserve GPU-backed VMs
- Fast access to premium hardware
- Best for solo developers and hands-on engineers
Flexible usage, no long-term commitment
Teams
Most popularFor startups and product teams moving from prototyping to reliable production inference.
Combine on-demand flexibility with reserved GPU capacity, predictable pricing, and production-ready infrastructure for growing workloads.
- On-demand plus reserved capacity
- Commitment discounts up to 35%
- Support for containers, clusters, and dedicated workloads
- Better cost control for recurring inference demand
- Best for teams shipping AI products
Scale usage without rebuilding your stack
Enterprise
For organizations that need dedicated infrastructure, custom deployment models, or platform-level control.
Deploy Istidlal as a managed platform, install it on-prem, or bring your own GPU devices and manage them through Istidlal.
- Dedicated infrastructure and large-scale reservations
- On-prem installation available
- Bring your own devices / GPUs
- Multi-month commitments and custom pricing
- Best for enterprise AI platforms and private environments
Built for large-scale and private deployments
Reserve your best price
NVIDIA H100 GPU
Receive discounted pricing for NVIDIA H100 GPUs with a commitment of hundreds of units.
- 80GB HBM3 Memory
- 3.35 TB/s Bandwidth
- PCIe Gen5 Support
NVIDIA H200 GPU
Access improved cost savings on NVIDIA H200 GPUs with a commitment of hundreds of units.
- 141GB HBM3e Memory
- 4.8 TB/s Bandwidth
- 40% Faster Inference
NVIDIA Blackwell Platform
Be among the first to get access to NVIDIA GB200 NVL72 and GB300 NVL72, the most advanced NVIDIA accelerators on the market.
- Liquid Cooled Rack
- 30X LLM Performance
- Multi-Node Scaling
NVIDIA GPU Instances
| GPU | vCPUs | RAM, GB | Price per GPU-hour |
|---|---|---|---|
| NVIDIA GB200 NVL72* | 112 | 800 | Contact us |
| NVIDIA HGX B200 | 16 | 200 | $5.50 |
| NVIDIA HGX H200 | 16 | 200 | $3.50 |
| NVIDIA HGX H100 | 16 | 200 | $2.95 |
| NVIDIA L40S GPU with AMD | 16-192 | 96-1152 | from $1.82 |
| NVIDIA L40S GPU with Intel | 8-40 | 32-160 | from $1.55 |
* GB200 NVL72 pricing available upon request. Contact our sales team for custom quotes.