Specifications Compared
| Spec | B200 SXM | RTX 5090 |
|---|---|---|
| TDP | 1000W | 575W |
| VRAM | 192 GB | 32 GB |
| CUDA Cores | 18,432 | 21,760 |
| FP4 (dense) | 9,000 TFLOPS | 1,676 TFLOPS |
| FP8 (dense) | 4,500 TFLOPS | 419 TFLOPS |
| Memory Type | HBM3e | GDDR7 |
| Architecture | Blackwell | Blackwell |
| FP16 (dense) | 2,250 TFLOPS | 209.5 TFLOPS |
| Form Factors | SXM | PCIe |
| INT8 (dense) | 4,500 TOPS | 838 TOPS |
| Interconnect | NVLink, PCIe 6.0, InfiniBand | PCIe 5.0 |
| Tensor Cores | 576 | 680 |
| FP32 Performance | 75 TFLOPS | 104.8 TFLOPS |
| FP64 Performance | 37 TFLOPS | 1.6 TFLOPS |
| Memory Bandwidth | 8,000 GB/s | 1,792 GB/s |
| FP4 (with sparsity) | 18,000 TFLOPS | 3,352 TFLOPS |
| FP8 (with sparsity) | 9,000 TFLOPS | 838 TFLOPS |
| FP16 (with sparsity) | 4,500 TFLOPS | 419 TFLOPS |
| INT8 (with sparsity) | 9,000 TOPS | 1,676 TOPS |
Performance Analysis
The B200 SXM variant records FP16 dense throughput at 2250 TFLOPS compared with the RTX 5090 variant at 209.5 TFLOPS while FP16 with sparsity reaches 4500 TFLOPS against 419 TFLOPS. FP32 performance stands at 75 TFLOPS for the B200 SXM variant versus 104.8 TFLOPS for the RTX 5090 variant. These dense to dense and sparsity to sparsity ratios indicate the B200 SXM variant sustains larger batch sizes during training workloads through its 8000 GB per second bandwidth whereas the RTX 5090 variant bandwidth of 1792 GB per second constrains batch capacity in memory intensive inference steps.
Current On-Demand Offers
Cheapest secure on-demand offer per provider that is reported in stock right now, per GPU per hour; multi-GPU instances show the whole-instance price alongside. Deploy opens the DeployGPU console for providers on the platform, otherwise the provider's own site. Stock is rechecked every minute.
B200 SXM
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | Deploy |
|---|---|---|---|---|---|
| RunPod | global | 1 | $7.99 | — | Deploy |
| VERDA | FIN-HEL | 8 | $8.03 | $64.23 | Deploy |
2 providers in stock, 2 offers. All B200 SXM offers, price history and alerts
Notify me when B200 SXM drops below a price
One email when the cheapest in-stock on-demand price per GPU-hour falls below your threshold. Cheapest right now: $7.99/GPU-hr.
QuantaCloud
Comparing B-series options? Get one quote for all of them.
Skip the per-provider sales calls. Reserved and cluster B-series configurations from 16 to 1024+ GPUs with InfiniBand fabric, 3 to 12 month terms. One quote at partner rates, 24h turnaround.
When to Choose the B200 SXM
The B200 SXM variant suits workloads that require 192 GB HBM3e capacity and 8000 GB per second bandwidth at 1000 W TDP. Its NVLink and InfiniBand interconnects support multi GPU scaling for large model training that exceeds the 32 GB limit of the alternative variant.
When to Choose the RTX 5090
The RTX 5090 variant fits tasks that operate within 32 GB GDDR7 capacity and 1792 GB per second bandwidth at 575 W TDP. Its PCIe 5.0 interface matches standard workstation deployments where the higher TDP and SXM form factor of the alternative variant add unnecessary overhead.
Use Cases
The B200 SXM variant supplies 192 GB HBM3e and 2250 TFLOPS FP16 dense throughput that accommodate models beyond the 32 GB capacity of the RTX 5090 variant.
The B200 SXM variant 8000 GB per second bandwidth supports larger concurrent batches than the 1792 GB per second available on the RTX 5090 variant.
The B200 SXM variant 4500 TFLOPS FP8 dense throughput and 192 GB memory enable fine tuning of models that exceed RTX 5090 variant limits.
The RTX 5090 variant 104.8 TFLOPS FP32 performance and 575 W TDP suit image generation workloads that fit inside 32 GB GDDR7.
The B200 SXM variant 75 TFLOPS FP32 rating combined with 192 GB capacity handles large simulation datasets beyond the RTX 5090 variant scope.
Frequently Asked Questions
What memory configurations distinguish the B200 SXM variant from the RTX 5090 variant?▾
The B200 SXM variant contains 192 GB HBM3e while the RTX 5090 variant contains 32 GB GDDR7. These capacities determine maximum model sizes each GPU can process.
How do the TDP values compare between the two variants?▾
The B200 SXM variant operates at 1000 W TDP whereas the RTX 5090 variant operates at 575 W TDP. System power and cooling designs must account for this difference.
Which interconnect options appear on each variant?▾
The B200 SXM variant includes NVLink, PCIe 6.0 and InfiniBand while the RTX 5090 variant uses PCIe 5.0. Multi node scaling therefore favors the B200 SXM variant.
What FP16 dense performance figures are listed for each GPU?▾
The B200 SXM variant lists 2250 TFLOPS FP16 dense and the RTX 5090 variant lists 209.5 TFLOPS FP16 dense. Direct comparison uses these matching dense values.
Do both variants share the same memory bandwidth?▾
No. The B200 SXM variant provides 8000 GB per second while the RTX 5090 variant provides 1792 GB per second. Bandwidth affects achievable batch sizes in both training and inference.
Which form factors are used by the compared variants?▾
The B200 SXM variant uses the SXM form factor and the RTX 5090 variant uses the PCIe form factor. Server chassis compatibility therefore differs between the two.
Which is cheaper to rent, the B200 SXM or the RTX 5090?▾
Cloud rental prices for both the B200 SXM and RTX 5090 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 B200 SXM have compared to the RTX 5090?▾
The B200 SXM has 192 GB of HBM3e memory. The RTX 5090 has 32 GB of GDDR7 memory.
Can I find B200 SXM and RTX 5090 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 B200 SXM and the RTX 5090?▾
The B200 SXM uses the Blackwell architecture (2024) while the RTX 5090 uses Blackwell (2025). The B200 SXM delivers 10.7x the dense FP16 throughput (2,250 vs 209.5 TFLOPS, both without sparsity) and 4.5x the memory bandwidth of the RTX 5090.
Rent these GPUs
Each GPU page lists every current on-demand offer by provider, per GPU-hour, updated every minute.
Related comparisons
How this page is made
- Specifications come from the NVIDIA, AMD and Intel datasheets for the B200 SXM and the RTX 5090. Dense and sparse throughput are listed separately.
- Prices and availability are live: every offer shown is an in-stock, on-demand listing from a provider we track, rechecked every minute.
- The written comparison (overview, performance notes, when to choose each card, verdict, use cases and FAQ) was drafted with an AI model from the specification table above and passed an automated check that rejects any figure not in that table. It was last generated on . It contains no prices; those are always read live.
- Read how we collect the data, or report an error on this page.
Next steps
- Rent B200Every current offer by provider, daily price history and a price alert.
- Rent RTX 5090Every current offer by provider, daily price history and a price alert.
- GPU price indexHow on-demand prices for the major GPUs have moved, updated daily.
- All GPUsEvery GPU we track, with current lows and provider counts.