Compare · GPUs & AI chips
Gaudi 3 vs Gaudi 2
RANKED BY DENSE FP16/BF16 TENSOR TFLOPS PER CHIP
#5 VS #14
CHECKED 06 SEPT 2026
Side by side
Verdict
Gaudi 3 ranks #5 of 20 on Dense FP16/BF16 tensor TFLOPS per chip (1.678 PFLOPS); Gaudi 2 ranks #14 (432 TFLOPS).
Rank 05 of 20 · leads
Dense FP16/BF16 tensor TFLOPS per chip
1.678 PFLOPS
MakerIntel
Date2024
Verified04 Sept 2026
Evidence
Intel Gaudi 3 AI Accelerator white paper (Intel document 817486), product-comparison table: "BF16 MME TFLOPS 1678" for Gaudi 3. The same paper's MME-precision table lists BF16 and FP8 at 1678 TFLOPS and "FP16 (signed)" at 459 TFLOPS — so this row is the BF16 matrix figure. 128 GB HBM2e. Not an FP16=BF16 part in the NVIDIA/AMD sense; the difference is disclosed here rather than hidden.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 Gaudi 3 matchups
Source Vendor product pages and official spec documents (AMD Instinct MI355X / MI350X / MI325X / MI300X / MI300A / MI250X / MI250 / MI210; NVIDIA…
Last checked 06 Sept 2026