NVIDIA H200 SXM
The H200 is NVIDIA's latest flagship AI accelerator, featuring 141GB of HBM3e memory for handling the largest language models. Built on Hopper architecture, it delivers exceptional performance for both training and inference workloads.
VRAM
141 GB
Memory
HBM3e
Bandwidth
4,800 GB/s
FP16
1,979 TF
TDP
700 W
Memory and compute
Architecture and power
Notes
Optimized for large language models with 141GB HBM3e
Models that may fit on NVIDIA H200 SXM
Estimates based on INT8 quantization. Actual fit depends on framework and batch size.
More GPUs by NVIDIA
The manufacturer's other cards in the catalog, largest memory first, with bandwidth and power where published.
Frequently asked questions about NVIDIA H200 SXM
Answered from the entry's own figures: memory, which models fit, bandwidth, power and the card's class.
NVIDIA H200 SXM has 141 GB of HBM3e at 4,800 GB/s. At INT8 a model's weights take about one byte per parameter, so the card holds roughly a 112B-parameter model with room for context; larger models need several cards or a lower precision.
From the catalog's estimates at INT8, Llama 3.2 90B Vision, Gemma 4 31B IT, Llama Nemotron Embed Vl 1B V2 and GLM Z1 9B 0414 fit on a single NVIDIA H200 SXM with their default context. The list on this page shows the estimated memory each takes; the capacity planner sizes any model against this card for your own context length, batch size and request rate.
4,800 GB/s. Bandwidth bounds how fast a model generates tokens, because every token reads the whole set of weights and the KV cache from memory; at this rate a 112B model at INT8 could be read on the order of 43 times a second on one card, before batching and framework efficiency.
NVIDIA H200 SXM has a TDP of 700 W, with a maximum of 700 W, in the SXM5 form factor. Budget the TDP per card plus host overhead when sizing a server's power and cooling.
NVIDIA H200 SXM is a datacenter accelerator, meant for servers and multi-GPU serving, built on the Hopper architecture and released in 2024. It supports a multi-GPU interconnect at 900 GB/s, so a model larger than one card can be split across several.
Specifications from the manufacturer's published material; a figure that is not published is shown as such, and a derived one is explained in the notes.
Added Jan 25, 2026 · Last updated Jan 25, 2026
Explore more
Companion tools that draw on the same catalog and routing engine.
Put intelligence in the middle.
One gateway in front of every model, with your policies applied and every decision on record. Start with $5 of credit and 5,000 routing decisions a month, no card required.