While the open LLM leaderboard provided (and still provides) a good starting point for a lot of companies exploring integrating LLMs into their products, the one-upmanship of it seems to be less and less of an indicator of whether a model is "good" for a specific product.

→ View original post on X — @mattlynley