Compare · GPUs & AI chips

Blackwell Ultra B300 SXM vs Gaudi 3

RANKED BY DENSE FP16/BF16 TENSOR TFLOPS PER CHIP #4 VS #5 CHECKED 06 SEPT 2026

Side by side

Verdict

Blackwell Ultra B300 SXM ranks #4 of 20 on Dense FP16/BF16 tensor TFLOPS per chip (2.25 PFLOPS); Gaudi 3 ranks #5 (1.678 PFLOPS).

Rank 04 of 20 · leads

Blackwell Ultra B300 SXM

Dense FP16/BF16 tensor TFLOPS per chip

2.25 PFLOPS

MakerNVIDIA

Date2025

Verified04 Sept 2026

Evidence

NVIDIA HGX Platform page, HGX B300 row ("8x NVIDIA Blackwell Ultra SXM", shipping now). FP16/BF16 Tensor Core is the same 36 PFLOPS sparse / 18 PFLOPS dense board total as B200, so 2.25 PFLOPS dense per chip. B300's advertised uplift versus B200 is at FP4 and HBM capacity, not at dense FP16/BF16 — so on this metric B200 and B300 are tied.

Rank 05 of 20

Gaudi 3

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.

Try another pair

More Blackwell Ultra B300 SXM 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

Full GPUs & AI chips ranking Spot an error? Send a tip