NVIDIA H100 NVL
The H100 NVL pairs two H100 GPUs connected via NVLink for 94GB combined memory per GPU. Optimized for large language model inference in PCIe-based systems.
VRAM
94 GB
Memory
HBM3
Bandwidth
3,900 GB/s
FP16
1,979 TF
TDP
400 W
Memory and compute
Architecture and power
Notes
Dual-GPU for inference
Models that may fit on NVIDIA H100 NVL
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 H100 NVL
Answered from the entry's own figures: memory, which models fit, bandwidth, power and the card's class.
NVIDIA H100 NVL has 94 GB of HBM3 at 3,900 GB/s. At INT8 a model's weights take about one byte per parameter, so the card holds roughly a 75B-parameter model with room for context; larger models need several cards or a lower precision.
From the catalog's estimates at INT8, Gemma 4 31B IT, Llama Nemotron Embed VL 1B V2, GLM Z1 9B 0414 and NVIDIA Nemotron Nano 12B V2 VL BF16 fit on a single NVIDIA H100 NVL 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.
3,900 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 75B model at INT8 could be read on the order of 52 times a second on one card, before batching and framework efficiency.
NVIDIA H100 NVL has a TDP of 400 W, with a maximum of 400 W, in the PCIe Dual-slot form factor. Budget the TDP per card plus host overhead when sizing a server's power and cooling.
NVIDIA H100 NVL is a datacenter accelerator, meant for servers and multi-GPU serving, built on the Hopper architecture and released in 2023. It supports a multi-GPU interconnect at 600 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 on the free plan with 5,000 requests a month, no card required.