Specifications Compared
| Spec | H200 | RTX-3090 |
|---|---|---|
| TDP | 700W | 350W |
| VRAM | 141 GB | 24 GB |
| CUDA Cores | 16,896 | 10,496 |
| Memory Type | HBM3e | GDDR6X |
| Architecture | Hopper | Ampere |
| Form Factors | SXM, NVL | PCIe |
| Interconnect | NVLink, PCIe 5.0, InfiniBand | NVLink |
| Tensor Cores | 528 | 328 |
| FP8 Performance | 3,958 TFLOPS | |
| FP16 Performance | 1,979 TFLOPS | 35.6 TFLOPS |
| FP32 Performance | 67 TFLOPS | 35.6 TFLOPS |
| FP64 Performance | 34 TFLOPS | |
| INT8 Performance | 3,958 TOPS | |
| Memory Bandwidth | 4,800 GB/s | 936 GB/s |
Performance Analysis
H200's FP16 performance of 1979 TFLOPS dwarfs RTX 3090's 35.6 TFLOPS, accelerating deep learning training by handling more operations per second: this shortens epochs for large datasets. FP32 at 67 TFLOPS on H200 versus 35.6 TFLOPS supports superior general-purpose computing and simulations. The FP16 to FP32 ratio on H200 favors tensor-heavy AI, while RTX 3090's parity suits mixed workloads.
Memory bandwidth defines practical limits: H200's 4800 GB/s enables massive batch sizes in training, fitting models into 141 GB VRAM without fragmentation, unlike RTX 3090's 936 GB/s and 24 GB constraining batches. For inference, H200's FP8 at 3958 TFLOPS boosts quantized LLM serving. TDP of 700W on H200 demands robust cooling, contrasting RTX 3090's efficient 350W for edge deployments.
Live Cloud Pricing
Real-time prices from 25+ providers. Updated every 60 seconds.
H200 NVL
| Provider | GPU Model | VRAM | Host Specs | Region | Price | Status | Action | |
|---|---|---|---|---|---|---|---|---|
 Vultr | NVIDIA GH200 Grace Hopper 96GB VRAM | 96GB | 72 vCPU 480GB RAM 960GB Storage | Atlanta | $1.99/GPU/hr | Available | ||
 Lambda Labs | NVIDIA GH200 Grace Hopper 96GB VRAM | 96GB | 64 vCPU 432GB RAM 4096GB Storage | Virginia | $2.29/GPU/hr | Available | ||
 Nebius | NVIDIA H200 SXM 141GB VRAM | 141GB | 16 vCPU 200GB RAM | 🌍Europe | $2.45/GPU/hr | |||
 CoreWeave | 8×NVIDIA H200 SXM 141GB VRAM | 141GB | 128 vCPU 0GB RAM 61440GB Storage | United States | $2.58/GPU/hr $20.64/hr total (8×) | |||
 Ori | 4×NVIDIA H200 SXM 141GB VRAM | 141GB | 96 vCPU 960GB RAM 12000GB Storage | London | $3.50/GPU/hr $14.00/hr total (4×) | Available |
RTX 3090
| Provider | GPU Model | VRAM | Host Specs | Region | Price | Status | Action | |
|---|---|---|---|---|---|---|---|---|
 TensorDock | NVIDIA GeForce RTX 3090 24GB VRAM | 24GB | 0 vCPU 0GB RAM | Wilmington, Delaware | $0.20/GPU/hr | Available | ||
TensorDock | NVIDIA GeForce RTX 3090 24GB VRAM | 24GB | 0 vCPU 0GB RAM | Dallas, Texas | $0.21/GPU/hr | Available | ||
 Vast.ai | 4×NVIDIA GeForce RTX 3090 24GB VRAM | 24GB | 32 vCPU 403GB RAM 153GB Storage | Iceland | $0.25/GPU/hr $1.01/hr total (4×) | Available | ||
Vast.ai | 4×NVIDIA GeForce RTX 3090 24GB VRAM | 24GB | 32 vCPU 252GB RAM 1440GB Storage | Finland | $0.27/GPU/hr $1.07/hr total (4×) | Available | ||
 LeaderGPU | 8×NVIDIA GeForce RTX 3090 24GB VRAM | 24GB | 64 vCPU 384GB RAM 2000GB Storage | Netherlands | $0.29/GPU/hr $2.29/hr total (8×) | Available |
When to Choose the H200 NVL
Enterprises training LLMs or running inference on models exceeding 24 GB select H200 NVL: its 141 GB VRAM accommodates full parameter sets, and 4800 GB/s bandwidth sustains high throughput. Datacenter interconnects like NVLink and PCIe 5.0 scale multi-GPU clusters seamlessly for production workloads.
When to Choose the RTX 3090
Developers prototyping or fine-tuning smaller models choose RTX 3090: 24 GB VRAM suffices for most tasks at $0.08/hr starting price. Its PCIe form factor and 350W TDP integrate easily into workstations, offering 35.6 TFLOPS FP16 for cost-sensitive experimentation.
Use Cases
H200's 141 GB VRAM and 1979 TFLOPS FP16 handle massive models and large batches. RTX 3090's 24 GB limits scale.
3958 TFLOPS FP8 and 4800 GB/s bandwidth on H200 enable high-throughput serving. 3090 struggles with models over 24 GB.
RTX 3090's 35.6 TFLOPS suffices for small models at low cost. H200 accelerates larger ones with 67 TFLOPS FP32.
RTX 3090's 24 GB GDDR6X and 936 GB/s bandwidth generate images efficiently at $0.08/hr. H200 overkill for consumer AI art.
H200's 67 TFLOPS FP32 and NVLink excel in simulations. RTX 3090's 35.6 TFLOPS fits lighter tasks.
Frequently Asked Questions
Which has more VRAM: H200 or RTX 3090?▾
H200 provides 141 GB HBM3e VRAM, vastly exceeding RTX 3090's 24 GB GDDR6X. This allows H200 to load enormous AI models single-GPU. RTX 3090 requires multi-GPU for similar capacity.
How do FP16 performances compare?▾
H200 achieves 1979 TFLOPS FP16, over 55 times RTX 3090's 35.6 TFLOPS. This gap accelerates AI training dramatically on H200. Inference benefits similarly from the disparity.
What is the price difference in cloud?▾
RTX 3090 starts at $0.08/hr averaging $0.44/hr across 44 offers. H200 NVL begins at $0.50/hr averaging $2.39/hr over 4 offers. Budget projects favor 3090.
Does H200 support FP8?▾
H200 delivers 3958 TFLOPS FP8 for quantized inference. RTX 3090 lacks native FP8 specs. This positions H200 for efficient LLM serving.
Which has higher memory bandwidth?▾
H200's 4800 GB/s outpaces RTX 3090's 936 GB/s by over 5 times. Larger batches and faster data movement result on H200. This impacts training throughput directly.
What are their TDPs?▾
H200 requires 700W TDP for peak performance. RTX 3090 uses 350W, easing power and cooling needs. Datacenter setups handle H200 readily.
Which is cheaper to rent, the H200 or the RTX 3090?▾
Cloud rental prices for both the H200 and RTX 3090 vary by provider, configuration, and availability. This page shows live pricing from 25+ providers updated every 60 seconds. Scroll to the Live Cloud Pricing section to compare current rates.
How much VRAM does the H200 have compared to the RTX 3090?▾
The H200 has 141 GB of HBM3e memory. The RTX 3090 has 24 GB of GDDR6X memory.
Can I find H200 and RTX 3090 GPUs available to rent right now?▾
Yes. This page shows real-time availability across 25+ cloud GPU providers. The Live Cloud Pricing section displays only in-stock offers with current pricing.
What is the main difference between the H200 and the RTX 3090?▾
The H200 uses the Hopper architecture (2024) while the RTX 3090 uses Ampere (2020). The H200 delivers 55.6x the FP16 throughput and 5.1x the memory bandwidth of the RTX 3090.