H200 vs A40
Explore a head to head comparison of specifications, performance, and pricing.
H200
The NVIDIA H200 is an advanced Hopper-based GPU that significantly boosts performance for generative AI, LLM, and HPC workloads with enhanced memory and bandwidth.
ManufacturerNVIDIA
GPU ArchitectureHopper
Average Price$14.99/hr
GPU VRAM141 GB
Cloud Availability7 clouds
System Memory2048 GB
CPU Cores480
Storage30.7 TB
A40
The NVIDIA A40 delivers high-performance computing capabilities for AI, machine learning, and data science applications.
ManufacturerNVIDIA
GPU Architecture—
Average Price$4.30/hr
GPU VRAM48 GB
Cloud Availability2 clouds
System Memory480 GB
CPU Cores96
Storage7.8 TB
See how the H200 & A40 compare
Compare detailed hardware specifications and average pricing for the H200 and A40.
Compare Hardware Specifications
| H200 | A40 | |
|---|---|---|
| GPU Type | H200 | A40 |
| VRAM per GPU | 141 GB | 48 GB |
| Manufacturer | NVIDIA | NVIDIA |
| Architecture | Hopper | Ampere |
| Interconnect | SXM5 | PCIe Gen4 |
| Memory Bandwidth | 4.8 TB/s | 696 GB/s |
| FP16 TFLOPS | 267.6 TFLOPS (4:1) | 37.42 TFLOPS (1:1) |
| CUDA Cores | 16896 | 10752 |
| Tensor Cores | 528 (4th Gen) | 336 (3rd Gen) |
| RT Cores | N/A | 84 (2nd Gen) |
| Base Clock | 1500 MHz | 1125 MHz |
| Boost Clock | 1980 MHz | 1740 MHz |
| TDP | 350-700W | 300W |
| Process Node | TSMC 4N | TSMC 8nm |
| Data Formats | FP8, INT8, BF16, FP16, TF32, FP32, FP64 | INT8, BF16, FP16, TF32, FP32 |
Compare Average On-Demand Pricing
| H200 | A40 | |
|---|---|---|
| 1 GPU | $3.44 /hr | $1.48 /hr |
| 2 GPUs | $6.82 /hr | $2.20 /hr |
| 4 GPUs | $12.00 /hr | $5.92 /hr |
| 8 GPUs | $22.56 /hr | $8.80 /hr |
Explore H200 & A40 Instances
Browse available instances with H200 and A40 GPUs. Filter by provider, availability, and more to find the perfect instance for your needs.
Explore more GPU comparisons
Select any two GPUs to compare their specifications and explore pricing across providers.
