NVIDIA H200 SXM 141GB
Performance
Peak theoretical throughput by precision type
| Precision | Dense | 2:4 Sparse |
|---|---|---|
FP64 64-bit floating point | 34TFLOPS | Structured sparsity is a tensor-core feature; this vector precision has no sparse form |
FP32 32-bit floating point | 67TFLOPS | Structured sparsity is a tensor-core feature; this vector precision has no sparse form |
TF32 TensorFloat-32 | 494TFLOPS | 989TFLOPS |
BF16 Brain Float 16 | 990TFLOPS | 1,979TFLOPS |
FP16 16-bit floating point | 990TFLOPS | 1,979TFLOPS |
FP8 8-bit floating point | 1,979TFLOPS | 3,958TFLOPS |
FP6 | No verified data available | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
FP4 | No verified data available | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
INT8 8-bit integer | 1,979TOPS | 3,958TOPS |
Dense peak divided by accelerator TDP. Board power only: excludes host CPUs, networking, cooling and facility overhead. TDP for this part: 700 W.
The H200 SXM 141GB in the GPU landscape
Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper
Higher and further left is better: more half-precision throughput for less power.
Specifications
Architecture
Hopper
Form Factor
SXM
Launch Year
2024
Process Node
4nm
Memory
141 GB
Bandwidth
4,800 GB/s
TDP
700 W
Max power (Flopper estimate)
~805 W est. Flopper estimate: 700 W TDP x 1.15. The vendor publishes no maximum board power for this part.
Spec Confidence
Official
Full Specifications
| Memory | |
|---|---|
| Memory | 141 GB |
| Memory Type | No verified data available |
| Bandwidth | 4.8 TB/s |
| Interface Width | No verified data available |
| Power & Thermal | |
| TDP | 700 W |
| Max power (Flopper estimate) | ~805 W est. |
| Enterprise Features | |
| Sparsity | Yes |
| General | |
| Form Factor | SXM |
| Architecture | Hopper |
| Process Node | 4nm |
| Launch Year | 2024 |
Datasheet & Resources
Flopper Datasheet
NVIDIA H200 SXM 141GB specifications, generated from our database. Printable.
Vendor product page
NVIDIA documentation for the H200 SXM 141GB
NVIDIA H200 Tensor Core GPU
NVIDIA · 2024-01-15
NVIDIA Developer Documentation
Technical specs, programming guides
Data Provenance
Every figure traced to a source
Primary Source
- Document
- NVIDIA H200 Tensor Core GPU
- Publisher
- NVIDIA
- Published
- 2024-01-15
Data Quality
- Spec confidence
- Official
- Clock basis
- Boost
- Core precisions with figures
- 7 of 9
- Normalization
- All values in TFLOPS
Rent this GPU
$2.09 / GPU-hr
Lowest of 54 live listings across 31 providers· median $4.47
Median of each provider's cheapest on-demand listing, so a provider with many configurations counts once.
Price history and every provider on the pricing page. Some links are affiliate links.
Similar GPUs
Systems Using This GPU
Pre-configured systems featuring the NVIDIA H200
| System | GPU Count | Peak Performance | Total Power | |
|---|---|---|---|---|
HPE ProLiant Compute XD685 (HGX H200) ProLiant Compute XD · rack · 2025 | 8x H200 | 15.8 PFLOPS | No verified data available | View System |
HPE Cray XD670 Cray · rack · 2024 | 8x H200 | 15.8 PFLOPS | No verified data available | View System |
NVIDIA DGX H200 DGX · rack · 2024 | 8x H200 | 15.8 PFLOPS | 8.5 kW | View System |
Gigabyte G593-SD1-LAX3 G593 · rack · 2024 | 8x H200 | 15.8 PFLOPS | No verified data available | View System |
Supermicro GPU SuperServer AS -8125GS-TNHR (HGX H200) GPU SuperServer · rack · 2024 | 8x H200 | 15.8 PFLOPS | No verified data available | View System |
Supermicro GPU SuperServer SYS-821GE-TNHR GPU SuperServer · rack · 2024 | 8x H200 | 15.8 PFLOPS | No verified data available | View System |
NVIDIA HGX H200 HGX · baseboard · 2024 | 8x H200 | 15.8 PFLOPS | 8.0 kW | View System |
Dell PowerEdge XE9680 PowerEdge · rack · 2024 | 8x H200 | 15.8 PFLOPS | No verified data available | View System |
Penguin Solutions Altus XE2318GTv2 Altus · rack · | 4x H200 | No verified data available | No verified data available | View System |
Penguin Solutions Altus XE5318GTSv2 Altus · rack · | 8x H200 | No verified data available | No verified data available | View System |
ASUS ESC N8-E11 ESC · rack · | 8x H200 | 15.8 PFLOPS | No verified data available | View System |
QCT QuantaGrid D74F-7U QuantaGrid · rack · | 8x H200 | No verified data available | No verified data available | View System |
QCT QuantaGrid D75F-7U QuantaGrid · rack · | 8x H200 | No verified data available | No verified data available | View System |
Penguin Solutions Relion XE3314GTS Relion · rack · | 4x H200 | No verified data available | No verified data available | View System |
Penguin Solutions Relion XE5318GTS Relion · rack · | 8x H200 | No verified data available | No verified data available | View System |
Frequently Asked Questions
How many TFLOPS does the NVIDIA H200 have?
The NVIDIA H200 delivers 67 TFLOPS FP32, 990 TFLOPS FP16 and 1,979 TFLOPS FP8 at peak.
What is the power consumption of the NVIDIA H200?
The NVIDIA H200 has a TDP (Thermal Design Power) rating of 700 watts.
How much memory does the NVIDIA H200 have?
The NVIDIA H200 is equipped with 141 GB of memory with 4,800 GB/s of memory bandwidth.
What architecture is the NVIDIA H200 based on?
The NVIDIA H200 is based on the Hopper architecture, launched in 2024.
How much does the NVIDIA H200 cost to rent?
The NVIDIA H200 rents from $2.09 per GPU-hour across 31 cloud providers, the lowest of 54 live listings Flopper tracks. The pricing page lists every provider and the price history.
When did the NVIDIA H200 come out?
The NVIDIA H200 launched in 2024 on the Hopper architecture.
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
No spam, unsubscribe anytime.