Jake Tech Lab

NVIDIA B300 vs AMD MI355X

AMD MI355X comes out ahead of NVIDIA B300 on 3 of the 6 things that decide this, going by what each maker publishes. The one that matters for you may be one the leader loses.

NVIDIA B300
NVIDIA B300
6.6
Fair
2 of 6 rounds
vs
AMD MI355XWinner
AMD MI355X
7.3
Good
3 of 6 rounds

The verdict

AMD MI355X takes it, 3–2 on rounds.

On the published figures, AMD MI355X takes 3 of 6 rounds. Treat that as a starting point and check the current figures on both pages, because prices and plans move.

Choose NVIDIA B300 if…

  • • You want NVIDIA B300's edge on how fast it runs fp4 ai math and how much power it draws

Choose AMD MI355X if…

  • • You want AMD MI355X's edge on how much memory it has and how fast its memory is

Round by round

  1. Round 1How much memory it hasAMD MI355X wins
    NVIDIA B3005/10

    270 GB HBM3E

    gpu memory

    • 270 GB HBM3E, against 288 GB HBM3E

    Source: dam-cdn.nvd.orangelogic.com ↗

    AMD MI355X8/10

    288 GB HBM3E

    dedicated memory size

    Source: amd.com ↗

    AMD MI355X wins this round on what the company itself publishes.

  2. Round 2How fast its memory isAMD MI355X wins
    NVIDIA B3005/10

    7.7 TB/s

    memory bandwidth

    • 7.7 TB/s, against 8 TB/s

    Source: dam-cdn.nvd.orangelogic.com ↗

    AMD MI355X8/10

    8 TB/s

    peak memory bandwidth

    Source: amd.com ↗

    AMD MI355X wins this round on what the company itself publishes.

  3. Round 3How fast it runs FP4 AI mathNVIDIA B300 wins
    NVIDIA B3008/10

    14 PFLOPS

    fp4 tensor core

    Source: dam-cdn.nvd.orangelogic.com ↗

    AMD MI355X5/10

    10.1 PFLOPs

    peak mxfp4 matrix

    • 10.1 PFLOPs, against 14 PFLOPS

    Source: amd.com ↗

    NVIDIA B300 wins this round on what the company itself publishes.

  4. Round 4How much power it drawsNVIDIA B300 wins
    NVIDIA B3008/10

    1,100 W

    max tdp

    Source: dam-cdn.nvd.orangelogic.com ↗

    AMD MI355X5/10

    1400W

    typical board power

    • 1400W, against 1,100 W

    Source: amd.com ↗

    NVIDIA B300 wins this round on what the company itself publishes.

  5. Round 5How it handles scientific FP64 mathAMD MI355X wins
    NVIDIA B3005/10

    1.2 TFLOPS

    fp64/fp64 tensor core

    • 1.2 TFLOPS, against 78.6 TFLOPs

    Source: dam-cdn.nvd.orangelogic.com ↗

    AMD MI355X8/10

    78.6 TFLOPs

    peak fp64 matrix

    Source: amd.com ↗

    AMD MI355X wins this round on what the company itself publishes.

  6. Round 6Which memory generation it usesTie
    NVIDIA B3007/10

    HBM3E

    memory type

    Source: dam-cdn.nvd.orangelogic.com ↗

    AMD MI355X7/10

    HBM3E

    dedicated memory type

    Source: amd.com ↗

    Nothing to separate them here.

NVIDIA B300NVIDIA B300
AMD MI355XAMD MI355X

Side by side

AMD MI355XBest for how fast its memory isAMD MI355X
Overall score
6.6
7.3
Best
Price
See price
See price
How fast its memory is
6.3
7.9
How much memory it has
7.3
8.0
How much power it draws
6.5
5.0
How fast it runs FP16 AI math
6.5
8.0
How new it is
8.0
7.8
Head-to-head findings
How fast its memory is
7.7 TB/s
8 TB/s
How much memory it has
270 GB HBM3E
288 GB HBM3E
How much power it draws
1,100 W
1400W
How fast it runs FP16 AI math
4.5 PFLOPS
2.5 PFLOPs
How new it is
2025
6/12/2025
Which memory generation it uses
HBM3E
HBM3E
How fast it runs FP8 AI math
9 PFLOPS
5 PFLOPs
How it handles scientific FP64 math
1.2 TFLOPS
78.6 TFLOPs
Record
Head-to-head record
1–2–1
7–0
Head-to-heads won
33 %
100 %
Rounds won
46 %
69 %
Pros
      Cons
      • 270 GB HBM3E, against 288 GB HBM3E
      • 7.7 TB/s, against 8 TB/s
      • 1.2 TFLOPS, against 78.6 TFLOPs
      • 9 PFLOPS, against 10 PFLOPS
      • 1400W, against 700 W
      • 1400W, against 750W
      • 78.6 TFLOPs, against 163.4 TFLOPs
      • 1400W, against 1000W
      Add a third option to this comparison →

      More AI accelerators match-ups