Will Groq's new LPUs overtake H100s in terms of inference at scale before EOY 2026? | Manifold

Will Groq's new LPUs overtake H100s in terms of inference at scale before EOY 2026?

19

Ṁ1kṀ1.5k

resolved Dec 9

Resolved

N/A

1H

6H

1D

1W

1M

ALL

Market context

Get

1,000

to start trading!

People are also trading

Will any of CPUs/GPU created by hackerfab generate at least 10^18 operations of compute by EOY2028?

Will any China-domestic AI chip reach ≥80% of NVIDIA H100 perf on a public benchmark before 2026-12-31?

Most training run compute greater than 2e27 FLOP by EOY 2026?

OpenAI reaches 250GW compute capacity by EOY 2033?

By EOY 2026, will it seem as if deep learning hit a wall by EOY 2025?

Company announces commercial room-temperature quantum processing unit (QPU) by EOY2027?

Will Jose Luis Ricon run more than 864 quintillion floating point ops on a NVIDIA h100/b200 (or better) by EOY2026

Before 2028, will a GPU the same or smaller die size as b100 achieve 2x or better max throughput on GPT-oss-120b?

What will happen to Groq in the next 3 years (before 2028)? [Suggest Options in comments]

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

Sort by:

Any details on the N/A?

Does this mean NVidia's successor to the H100 vs Groq' successor to the LPU?

@HastingsGreer Need clarification on this. The h100 may be surpassed by a new LPU, but I doubt the current gen LPU will beat what NVIDIA comes up with between now and market close

@ConnorTabarrok Talking about current-gen LPUs. Fresh released units. Not what either corp will offer in the future.

People are also trading

Will any of CPUs/GPU created by hackerfab generate at least 10^18 operations of compute by EOY2028?

Will any China-domestic AI chip reach ≥80% of NVIDIA H100 perf on a public benchmark before 2026-12-31?

Most training run compute greater than 2e27 FLOP by EOY 2026?

OpenAI reaches 250GW compute capacity by EOY 2033?

By EOY 2026, will it seem as if deep learning hit a wall by EOY 2025?

Company announces commercial room-temperature quantum processing unit (QPU) by EOY2027?

Will Jose Luis Ricon run more than 864 quintillion floating point ops on a NVIDIA h100/b200 (or better) by EOY2026

Before 2028, will a GPU the same or smaller die size as b100 achieve 2x or better max throughput on GPT-oss-120b?

What will happen to Groq in the next 3 years (before 2028)? [Suggest Options in comments]

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

Related questions

Will any of CPUs/GPU created by hackerfab generate at least 10^18 operations of compute by EOY2028?

Will any China-domestic AI chip reach ≥80% of NVIDIA H100 perf on a public benchmark before 2026-12-31?

Most training run compute greater than 2e27 FLOP by EOY 2026?

OpenAI reaches 250GW compute capacity by EOY 2033?

By EOY 2026, will it seem as if deep learning hit a wall by EOY 2025?

Company announces commercial room-temperature quantum processing unit (QPU) by EOY2027?

Will Jose Luis Ricon run more than 864 quintillion floating point ops on a NVIDIA h100/b200 (or better) by EOY2026

Before 2028, will a GPU the same or smaller die size as b100 achieve 2x or better max throughput on GPT-oss-120b?

What will happen to Groq in the next 3 years (before 2028)? [Suggest Options in comments]

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