What will be the maximum achievable flop utilization on the next generation of Nvidia server chips? | Manifold

What will be the maximum achievable flop utilization on the next generation of Nvidia server chips?

35

1.7kṀ1431

Dec 29

1%

<30%

1.5%

30-40%

6%

40-50%

9%

50-60%

34%

60-70%

23%

70-80%

11%

80-90%

14%

90-100%

Concretely, what is the best FLOPS (Floating Point Operations per Second) the next generation of Nvidia server cards will be able to achieve on fp16 matrix multiplications on matrices generated by the normal distribution, divided by the maximum theoretical FLOPS that Nvidia reports?

For example, for A100s, it's possible to achieve about 280+ TeraFLOPS out of a maximum of 312 TeraFLOPS, for a maximum flop utilization of ~90%.

On H100s, it seems to be closer around 700 TeraFLOPS, out of a maximum of 1000.

Will resolve when values seem clear after HNext cards are released, or a maximum of one year after Nvidia announces it.

To clarify, this will be for the B100, not the B200.

Get

1,000

to start trading!

Sort by:

I wouldn’t have a means to test this, but I wonder if the answer could be over 100% using liquid nitrogen and heavy overclocking.

To clarify, I will be basing this off the standard configuration (i.e. the listed 700W in their spec). If Nvidia sells an unusual spec with a higher power limit, I won't be using that to resolve the market.

Related questions

Will AI accelerators improve in FLOPs/watt by 100x of an NVidia H100 by 2033?

Will NVIDIA maintain a >=75% of the Data center market share for at least 4 quarters by the end of 2025?

If China invades Taiwan in 2023-2030, what will FLOP/s per dollar of top-ML GPUs be 10 years later?

If China invades Taiwan in 2023-2030, what will FLOP/s per dollar of top-ML GPUs be 5 years later?

Will Nvidia still retain over 50% market share of PC gamer GPU usage by 2030

Will there be an announcement of a model with a training compute of over 1e30 FLOPs by the end of 2025?

Will NVIDIA maintain a >=75% of the Data center market share for at least 8 quarters by the end of 2026?

Will the Groq chip inspire Nvidia/AMD to produce radically new AI chips before 2026?

Will Nvidia report gross margins lower than 60% in any year by 2028?

Will a machine learning training run exceed 10^26 FLOP in China before 2026?

Related questions

Will AI accelerators improve in FLOPs/watt by 100x of an NVidia H100 by 2033?

Will there be an announcement of a model with a training compute of over 1e30 FLOPs by the end of 2025?

Will NVIDIA maintain a >=75% of the Data center market share for at least 4 quarters by the end of 2025?

Will NVIDIA maintain a >=75% of the Data center market share for at least 8 quarters by the end of 2026?

If China invades Taiwan in 2023-2030, what will FLOP/s per dollar of top-ML GPUs be 10 years later?

Will the Groq chip inspire Nvidia/AMD to produce radically new AI chips before 2026?

If China invades Taiwan in 2023-2030, what will FLOP/s per dollar of top-ML GPUs be 5 years later?

Will Nvidia report gross margins lower than 60% in any year by 2028?

Will Nvidia still retain over 50% market share of PC gamer GPU usage by 2030

Will a machine learning training run exceed 10^26 FLOP in China before 2026?

© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules