If you’re curious, you can check the progress of many open source LLM’s and how they perform on various evals here:

https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderb...

With proprietary models: https://chat.lmsys.org/?leaderboard

I think we are only at the beginning. Call it first generation. LLaMA was even missing many improvements that existed before it (xPos, Multi-Query Attention, Blockwise Parallel Transformer)

Many researchers are improving very fast, and I would bet that soon we will see more efficient LLMs.

I don't think we need anymore high-flying search engines, do we?

salesforce air esearch ;)

We’re still in the early innings.

Long way from that

In terms of Gartner hype cycle, we are at the peak of inflated expectations for llms I reckon.

When it's superhuman.

To quote Dennis Reynolds, it hasn't even begun to peak.

How far we've come from GPT2...

Biggest models crossed the threshold between glorified Markov chain and proto-intelligence, so sure enough, expectations shoot up.