[DenseComet]

Interesting to see that their model comparisons don’t include OpenAI models.

[CuriouslyC]

Maybe not the reason, but claude sonnet obliterates gpt3.5 and there isn't a direct llama competitor to gpt4.

[gs17]

The 400B model seems to be a competitor, maybe not in parameter count, but benchmark-wise it seems to be similar.