H200
H200 is a NVIDIA GPU based on the Hopper SM90 architecture, first released in 2024.
Built from BenchGecko data as of April 14, 2026 · updates when the data changes
H200 is a NVIDIA GPU based on the Hopper SM90 architecture, first released in 2024.
Basic
The NVIDIA H200 is an AI GPU made by NVIDIA, codenamed Hopper. Peak throughput is 989 TFLOPS at FP16/BF16 and 1,979 TFLOPS at FP8. It carries 141 GB of HBM3e with 4.8 TB/s of memory bandwidth. Rated power (TDP) is 700 W. It is built on N4 at TSMC.
Deep
In the BenchGecko dataset it sits in the mainstream tier: current-generation non-flagship parts or last-generation flagships still widely deployed. It follows the H100. Disclosed buyers include Amazon. Specs come from manufacturer datasheets; the spec card on this page shows the dataset values and their as-of date.
Expert
That is 1.41 FP16 TFLOPS per watt. Memory bandwidth per unit of compute is 4.85 GB/s per FP16 TFLOP: a higher ratio helps memory-bound inference, while peak TFLOPS matter more for compute-bound training. Fabrication: Taiwan. Real throughput depends on the model, precision, batch size and software stack, so compare tokens per second per dollar for your own workload.
Depending on why you're here
- ·H200: 989 TFLOPS FP16/BF16 · 1,979 TFLOPS FP8
- ·Process: N4 at TSMC
- ·Tier: mainstream
- ·141 GB of HBM3e per chip
- ·Cloud availability in the dataset: AWS, Lambda and CoreWeave
- ·Compare it with other chips on /hardware/h200
- ·Disclosed buyers: Amazon
- ·NVIDIA · fabbed at TSMC
- ·Supply tightness is tracked on /hardware
- ·H200 is a chip built to run AI, made by NVIDIA
- ·It has 141 GB of fast memory so large models fit
- ·Companies buy or rent these to train and serve AI models
Frequently Asked Questions
Read the primary sources
- NVIDIA H200 product pagewww.nvidia.com