RTX 5090 vs H100 SXM
H100 SXM has 2.5× the memory; H100 SXM has 1.9× the bandwidth; RTX 5090 rents for 3.5× less.
The NVIDIA RTX 5090 (Blackwell, 2025) and the NVIDIA H100 SXM (Hopper, 2022) are an odd couple on paper, the H100 SXM is datacenter silicon and the RTX 5090 is a consumer flagship, and the price gap is exactly why people cross-shop them. On capacity, the H100 SXM carries 2.5x the memory (80 vs 32 GB), which sets what each can hold at all. On memory bandwidth, the number that governs LLM serving speed, the H100 SXM leads at 3.35 vs 1.79 TB/s. Raw compute favors the H100 SXM by 4.7x, which matters for training, prefill-heavy traffic, and diffusion work more than for chat-style decoding.
Price is where it settles: the RTX 5090 rents from $0.54/hr against $1.90/hr for the H100 SXM, a 3.5x gap that the performance numbers above only partly close. Per gigabyte of VRAM, the RTX 5090 is the cheaper rental, worth knowing if your model is capacity-bound rather than speed-bound. Bear in mind the RTX 5090 is a consumer card rented mostly through marketplaces, with the trust and reliability trade-offs that implies, while the H100 SXM is datacenter silicon with ECC HBM and standard hosting. For how these numbers translate into tokens per dollar, the inference cost estimator runs both cards against any model. Everything below is the underlying data.
NVIDIA · Blackwell · 2025
NVIDIA RTX 5090
The fastest thing you can put in a desktop: 32 GB of GDDR7 at 1.8 TB/s.
From $0.54/hr at Vast.ai
NVIDIA · Hopper · 2022
NVIDIA H100 SXM
The workhorse of the AI boom. Still the most widely available serious training and inference GPU.
From $1.90/hr at Hyperstack
Head to head
■ RTX 5090 ■ H100 SXM, bars share one scale across the whole catalog.
| RTX 5090 | H100 SXM | |
|---|---|---|
| Memory | 32 GB GDDR7 | 80 GB HBM3 |
| Bandwidth | 1.79 TB/s | 3.35 TB/s |
| FP16 dense | 210 TF | 990 TF |
| FP8 dense | 420 TF | 1,979 TF |
| TDP | 575 W | 700 W |
| Interconnect | PCIe Gen5 | NVLink 4 · 900 GB/s |
| Cheapest rental | $0.54/hr | $1.90/hr |
| $/hr per GB VRAM | $1.7¢ | $2.4¢ |
What one can run that the other can't
Only on RTX 5090
Nothing, H100 SXM runs everything RTX 5090 does.
Only on H100 SXM
- Llama 3.3 70B (4-bit)
- Qwen2.5 72B (4-bit)
- Mixtral 8x7B (8-bit)
Rule of thumb: for LLM serving, prefer the GPU with more memory bandwidth per dollar; for training and prefill-heavy work, prefer FLOPS per dollar; and if the model doesn't fit in VRAM, none of the other numbers matter. Sanity-check with the VRAM calculator.