GPT-4 performance and compute efficiency from a simple architecture before 2026

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

In what year will a GPT4-equivalent model be able to run on consumer hardware?

In what year will a GPT4-equivalent model be able to run on consumer hardware?

By 2030, will China have reached GPU parity with the US?

Conditional on no Doom by 2030, will China have reached GPU parity with the US?

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

Which, if any, GPT-n will outperform AlphaGeometry merely via prompting, by 2030?

Will the performance jump from GPT4->GPT5 be less than the one from GPT3->GPT4?

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