NVIDIA A100 SXM (40GB) vs NVIDIA H200 SXM — GPU Comparison (Apr 2026)
NVIDIA A100 SXM (40GB) (40GB HBM2e, 312 TFLOPS FP16, Ampere) vs NVIDIA H200 SXM (141GB HBM3e, 990 TFLOPS FP16, Hopper). Cloud pricing: NVIDIA A100 SXM (40GB) from $0.80/hr, NVIDIA H200 SXM from $2.05/hr. Compare specs, VRAM, performance, and pricing across 3 cloud providers to find the best GPU for your AI workload.
|
NVIDIA A100 SXM (40GB)
40GB HBM2e · Ampere
|
NVIDIA H200 SXM
141GB HBM3e · Hopper
|
||
|---|---|---|---|
| Specifications | |||
| Manufacturer | NVIDIA | NVIDIA | |
| Architecture | Ampere | Hopper | |
| VRAM | 40 GB HBM2e | 141 GB HBM3e | |
| Memory Bandwidth | 1,555 GB/s | 4,800 GB/s | |
| FP16 (Tensor) | 312.0 TFLOPS | 990.0 TFLOPS | |
| FP32 | 19.5 TFLOPS | 67.0 TFLOPS | |
| TDP | 400W | 700W | |
| Release Year | 2020 | 2024 | |
| Segment | Data center | Data center | |
| Best For | AI training fine-tuning inference on smaller models | Large model training inference for 70B+ parameter models | |
| Cloud Pricing | |||
| Cheapest On-Demand | $0.80/hr | $2.05/hr | |
| Cheapest Spot | — | $3.59/hr | |
| Providers | 2 | 3 | |
| Provider Pricing (On-Demand) | |||
|
$0.80/hr | $2.05/hr | |
|
$1.14/hr | $3.59/hr | |
|
N/A | $3.44/hr | |
NVIDIA A100 SXM (40GB)
NVIDIA H200 SXM
Related GPU Comparisons
NVIDIA A100 SXM (40GB) vs NVIDIA H100 SXMNVIDIA H200 SXM vs NVIDIA H100 SXMNVIDIA A100 SXM (40GB) vs NVIDIA A100 SXM (80GB)NVIDIA H200 SXM vs NVIDIA A100 SXM (80GB)NVIDIA A100 SXM (40GB) vs NVIDIA L40SNVIDIA H200 SXM vs NVIDIA L40SNVIDIA A100 SXM (40GB) vs NVIDIA GeForce RTX 4090NVIDIA H200 SXM vs NVIDIA GeForce RTX 4090
Top Providers for NVIDIA A100 SXM (40GB) and NVIDIA H200 SXM
These 3 providers offer both NVIDIA A100 SXM (40GB) and NVIDIA H200 SXM. Full head-to-head comparison of GPU models, pricing, infrastructure, and developer tools.
Vast.ai vs RunPod vs DigitalOcean - GPU Provider Comparison (April 2026)
Side-by-side comparison of Vast.ai vs RunPod vs DigitalOcean. Quickly compare GPU models, hourly pricing, spot instances, billing granularity, VRAM, regions, developer tools, Kubernetes support, and compliance to narrow down your cloud GPU provider shortlist. Data updated April 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.
|
DigitalOcean
Simple, scalable GPU cloud for AI/ML
|
|
|---|---|---|---|
| Overview | |||
| Trustpilot Rating | 4.4 | 3.7 | 4.6 |
| Headquarters | United States | United States | United States |
| Provider Type | GPU Marketplace | GPU-Focused | N/A |
| 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 | AI training inference fine-tuning LLM deployment LLM serving computer vision startups generative AI research |
| 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 | RTX 4000 Ada RTX 6000 Ada L40S MI300X H100 SXM H200 |
| Max VRAM (GB) | 192 | 288 | 192 |
| Max GPUs/Instance | 8 | 8 | 8 |
| Interconnect | NVLink, InfiniBand | NVLink | NVLink |
| Pricing | |||
| Starting Price ($/hr) | $0.06/hr | $0.06/hr | $0.76/hr |
| Billing Granularity | Per-second | Per-second | Per-second |
| Spot/Preemptible | Yes | Yes | No |
| Reserved Discounts | Up to 50% (1-6 month reserved) | 15-29% (1-month to 1-year plans) | N/A |
| Free Credits | Small test credit on signup | $5-$500 bonus after first $10 spend | $200 free credit for 60 days |
| Egress Fees | Varies by host ($/TB) | None (Free) | None (included in plan) |
| 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) | 500-720 GiB NVMe boot (included), 5 TiB NVMe scratch on larger configs, Volumes at $0.10/GiB/mo |
| Infrastructure | |||
| Regions | 500+ locations, 40+ data centers | 31 global regions | New York (NYC2), Toronto (TOR1), Atlanta (ATL1), Richmond (RIC1), Amsterdam (AMS3) |
| Uptime SLA | No formal SLA (host reliability scores visible) | 99.99% | 99% |
| Developer Experience | |||
| Frameworks | PyTorch TensorFlow CUDA vLLM ComfyUI | PyTorch TensorFlow JAX ONNX CUDA | PyTorch TensorFlow Jupyter Miniconda CUDA ROCm Hugging Face |
| Docker Support | Yes | Yes | Yes |
| SSH Access | Yes | Yes | Yes |
| Jupyter Notebooks | Yes | Yes | Yes |
| API / CLI | Yes | Yes | Yes |
| Setup Time | Seconds | Instant | Minutes |
| Kubernetes Support | No | No | Yes |
| Business Terms | |||
| Min Commitment | None | None | None |
| Compliance | SOC 2 Type 2 HIPAA GDPR CCPA | SOC 2 Type II | SOC 2 Type II SOC 3 HIPAA (with BAA) CSA STAR Level 1 |
Vast.ai
RunPod
DigitalOcean
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.