Compare · GPUs & AI chips
H100 SXM vs Gaudi 2
RANKED BY DENSE FP16/BF16 TENSOR TFLOPS PER CHIP
#9 VS #14
CHECKED 06 SEPT 2026
Side by side
Verdict
H100 SXM ranks #9 of 20 on Dense FP16/BF16 tensor TFLOPS per chip (0.990 PFLOPS); Gaudi 2 ranks #14 (432 TFLOPS).
Rank 09 of 20 · leads
Dense FP16/BF16 tensor TFLOPS per chip
0.990 PFLOPS
MakerNVIDIA
Date2022
Verified04 Sept 2026
Evidence
NVIDIA H100 product page, H100 SXM column: FP16/BF16 Tensor Core 1,979 teraFLOPS. H100 datasheet footnote 2: "With sparsity." Dense is 1,979 / 2 = 989.5 TFLOPS (0.990 PFLOPS). 80 GB HBM3. Tied with H200 SXM on this compute metric.Rank 14 of 20
Dense FP16/BF16 tensor TFLOPS per chip
432 TFLOPS
MakerIntel
Date2022
Verified04 Sept 2026
Evidence
Intel Gaudi 3 white paper (document 817486) comparison table lists Gaudi 2 "BF16 MME TFLOPS 432" and "FP8 MME TFLOPS 865". This row uses the BF16 MME cell. 96 GB HBM2e. The same comparison table is the official Intel source already used for the Gaudi 3 row.Try another pair
More H100 SXM matchups
- H100 SXM vs Instinct MI355X
- H100 SXM vs Instinct MI350X
- H100 SXM vs Blackwell B200 SXM
- H100 SXM vs Blackwell Ultra B300 SXM
- H100 SXM vs Gaudi 3
- H100 SXM vs Instinct MI325X
- H100 SXM vs Instinct MI300X
- H100 SXM vs H200 SXM
- H100 SXM vs Instinct MI300A
- H100 SXM vs TPU v6e (Trillium)
- H100 SXM vs Trainium2
- H100 SXM vs TPU v5p
Source Vendor product pages and official spec documents (AMD Instinct MI355X / MI350X / MI325X / MI300X / MI300A / MI250X / MI250 / MI210; NVIDIA…
Last checked 06 Sept 2026