# H100 SXM5 94 GB

> NVIDIA · Hopper

Canonical: https://gpuninja.com/chips/h100-sxm5-94-gb
Data as of: 2026-10-11T17:24:51Z
Path: Home › GPU chips
- Specs verified: between 11 Oct 2026, 17:23 UTC and 11 Oct 2026, 17:24 UTC

## Specifications

- Vendor: NVIDIA (Verified · source docs.nvidia.com · retrieved 11 Oct 2026, 17:24 UTC)
- Architecture: Hopper (Verified · source docs.nvidia.com · retrieved 11 Oct 2026, 17:24 UTC)
- Die: GH100 (Verified · source pci-ids.ucw.cz · retrieved 11 Oct 2026, 16:54 UTC · checked 11 Oct 2026, 17:23 UTC)
- Process: TSMC 4N (Verified · source ece.lsu.edu · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:24 UTC)
- Shader cores: 16,896 (Verified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)
- Tensor cores: 528 (Verified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)
- RT cores: None
- Memory bus: Unknown
- Max VRAM: 94 GB (Verified · source docs.nvidia.com · retrieved 11 Oct 2026, 17:24 UTC)
- Reference boost clock: Unknown
- Reference TDP: 700 W (Verified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)
- Host interface: Not recorded
- Launch MSRP: No public MSRP
- Launch date: Unknown
- Segment: Data center (Verified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)
- PCI IDs: 10de:2339 (Verified · source pci-ids.ucw.cz · retrieved 11 Oct 2026, 16:54 UTC · checked 11 Oct 2026, 17:23 UTC)
- Status: Active (Unverified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)

## Performance

- Memory bandwidth: 2,400 GB/s (Verified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)
- FP32 compute: 67 TFLOPS (Verified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)
- FP16 compute: 989.5 TFLOPS (Unverified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)
- BF16 compute: 989.5 TFLOPS (Unverified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)
- FP8 compute: 1,979 TFLOPS (Unverified · source ece.lsu.edu · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:24 UTC)
- FP4 compute: Not supported
- INT8 compute: 1,979 TOPS (Unverified · source lenovopress.lenovo.com · retrieved 11 Oct 2026, 17:01 UTC · checked 11 Oct 2026, 17:23 UTC)

## Boards built on this chip

Board, Chip, VRAM, TDP, Length, Slots, Price, Stock

No boards recorded for this chip.

## Benchmarks

No benchmark results recorded.

## Sources

- https://docs.nvidia.com/datacenter/tesla/mig-user-guide/latest/supported-gpus.html — Vendor, Architecture, Max VRAM; last verified 11 Oct 2026, 17:24 UTC
- https://pci-ids.ucw.cz/v2.2/pci.ids — Die, PCI IDs; last verified 11 Oct 2026, 17:23 UTC
- https://www.ece.lsu.edu/koppel/gp/refs/gtc22-whitepaper-hopper.pdf — Process, FP8 compute; last verified 11 Oct 2026, 17:24 UTC
- https://lenovopress.lenovo.com/lp1732 — Shader cores, Tensor cores, Reference TDP, Segment, Status, Memory bandwidth, FP32 compute, FP16 compute, BF16 compute, INT8 compute; last verified 11 Oct 2026, 17:23 UTC