GPU Providers
HQ
Billing
Actions
Cloud GPU Providers with Spot / Preemptible Instances
Spot or preemptible GPU instances offer 50-90% savings compared to on-demand pricing, in exchange for the possibility of interruption during high-demand periods. They are ideal for fault-tolerant workloads like distributed training with checkpointing, batch inference, and hyperparameter sweeps. This guide lists cloud GPU providers that offer spot pricing, helping you significantly reduce your GPU compute costs.
Trustpilot Reviews
226
+5
(7d)
+13
(30d)
HQ
United States Starting Price
$0.06/hr
Max VRAM
192 GB
Max GPUs
8
Billing
Per-second
Trustpilot Reviews
224
+4
(7d)
+13
(30d)
HQ
United States Starting Price
$0.06/hr
Max VRAM
288 GB
Max GPUs
8
Billing
Per-second
Trustpilot Reviews
6
+1
(7d)
+1
(30d)
HQ
United States Starting Price
$0.11/hr
Max VRAM
80 GB
Max GPUs
8
Billing
Per-second
Trustpilot Reviews
550
+2
(7d)
+9
(30d)
HQ
United States Starting Price
$0.47/hr
Max VRAM
288 GB
Max GPUs
16
Billing
Per-hour
Vast.ai vs RunPod - Comparison of Top Firms in This Guide
Vast.ai vs RunPod - GPU Provider Comparison (May 2026)
Head-to-head comparison of Vast.ai and RunPod. Compare GPU models, hourly pricing, billing granularity, spot instances, VRAM, infrastructure, developer tools, Kubernetes support, and compliance before choosing a provider. Data refreshed May 2026.
|
Vast.ai
Instant GPUs. Transparent Pricing.
|
RunPod
The cloud built for AI — deploy and scale GPU workloads from serverless inference to instant multi-node clusters on demand.
|
|
|---|---|---|
| Overview | ||
| Trustpilot Rating | 4.3 | 3.5 |
| Headquarters | United States | United States |
| Provider Type | GPU Marketplace | GPU-Focused |
| Best For | AI training inference fine-tuning Stable Diffusion batch processing research LLM serving generative AI | AI training inference fine-tuning Stable Diffusion batch processing rendering research LLM serving generative AI |
| GPU Hardware | ||
| GPU Models | B200 H200 H100 SXM H100 NVL A100 SXM A100 PCIe RTX 5090 RTX 5080 RTX 5070 Ti RTX 6000 Pro RTX 6000 Ada RTX 4500 Ada RTX A6000 RTX A5000 RTX A4000 L40S L40 A40 A10 RTX 4090 RTX 4080 RTX 4070 Ti RTX 4070 RTX 4060 Ti RTX 4060 RTX 3090 Ti RTX 3090 RTX 3080 Ti RTX 3080 RTX 3070 Ti RTX 3070 Tesla V100 Tesla T4 A2 GTX 1080 | B300 B200 H200 H100 SXM H100 PCIe H100 NVL MI300X A100 SXM A100 PCIe RTX 5090 RTX PRO 6000 L40S L40 RTX 6000 Ada RTX 5000 Ada RTX A6000 RTX A5000 RTX 4090 RTX 4080 SUPER RTX 4080 RTX 4070 Ti RTX 3090 Ti RTX 3090 RTX 3080 Ti RTX 3080 RTX 3070 A40 A30 A2 L4 |
| Max VRAM (GB) | 192 | 288 |
| Max GPUs/Instance | 8 | 8 |
| Interconnect | NVLink, InfiniBand | NVLink |
| Pricing | ||
| Starting Price ($/hr) | $0.06/hr | $0.06/hr |
| Billing Granularity | Per-second | Per-second |
| Spot/Preemptible | Yes | Yes |
| Reserved Discounts | Up to 50% (1-6 month reserved) | 15-29% (1-month to 1-year plans) |
| Free Credits | Small test credit on signup | $5-$500 bonus after first $10 spend |
| Egress Fees | Varies by host ($/TB) | None (Free) |
| Storage | Varies by host ($/GB/hr, charged while instance exists) | Container/Volume ($0.10/GB/mo), Idle Volume ($0.20/GB/mo), Network Storage ($0.07/GB/mo 1TB) |
| Infrastructure | ||
| Regions | 500+ locations, 40+ data centers | 31 global regions |
| Uptime SLA | No formal SLA (host reliability scores visible) | 99.99% |
| Developer Experience | ||
| Frameworks | PyTorch TensorFlow CUDA vLLM ComfyUI | PyTorch TensorFlow JAX ONNX CUDA |
| Docker Support | Yes | Yes |
| SSH Access | Yes | Yes |
| Jupyter Notebooks | Yes | Yes |
| API / CLI | Yes | Yes |
| Setup Time | Seconds | Instant |
| Kubernetes Support | No | No |
| Business Terms | ||
| Min Commitment | None | None |
| Compliance | SOC 2 Type 2 HIPAA GDPR CCPA | SOC 2 Type II |
Vast.ai
RunPod
Build your own comparison
Select any 2-6 firms from this guide and open them in the full comparison table.
Tip: if you do not select any firms we will start with the top 2 from this guide.