GH200 vs RTX4090
Explore a head to head comparison of specifications, performance, and pricing.
GH200
The NVIDIA GH200 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$2.86/hr
GPU VRAM96 GB
Cloud Availability2 clouds
System Memory480 GB
CPU Cores144
Storage4.8 TB
RTX4090
The nvidia RTX4090 delivers high-performance computing capabilities for AI, machine learning, and data science applications.
Manufacturernvidia
GPU Architecture—
Average Price$2.69/hr
GPU VRAM24 GB
Cloud Availability3 clouds
System Memory706 GB
CPU Cores120
Storage15.4 TB
See how the GH200 & RTX4090 compare
Compare detailed hardware specifications and average pricing for the GH200 and RTX4090.
Compare Hardware Specifications
| GH200 | RTX4090 | |
|---|---|---|
| GPU Type | GH200 | RTX4090 |
| VRAM per GPU | 96 GB | 24 GB |
| Manufacturer | NVIDIA | nvidia |
| Architecture | Hopper | N/A |
| Interconnect | NVLink-C2C | pcie |
| Memory Bandwidth | 4 TB/s or 4.9 TB/s | N/A |
| FP16 TFLOPS | 267.6 TFLOPS (4:1) | N/A |
| CUDA Cores | 16896 | N/A |
| Tensor Cores | 528 (4th Gen) | N/A |
| Base Clock | 1500 MHz | N/A |
| Boost Clock | 1980 MHz | N/A |
| TDP | 900W-1000W | N/A |
| Process Node | TSMC 4N | N/A |
| Data Formats | FP8, INT8, BF16, FP16, TF32, FP32, FP64 | N/A |
Compare Average On-Demand Pricing
| GH200 | RTX4090 | |
|---|---|---|
| 1 GPU | $2.86 /hr | $0.60 /hr |
| 2 GPUs | N/A | $1.20 /hr |
| 4 GPUs | N/A | $2.40 /hr |
| 8 GPUs | N/A | $3.97 /hr |
Explore GH200 & RTX4090 Instances
Browse available instances with GH200 and RTX4090 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.
