News for hackers. Stuff that matters.

A comment on Google, Nvidia, and OpenAI – Stratechery by Ben Thompson. Parent comment

Re:Google, Nvidia, and OpenAI – Stratechery by Ben Thompson by cj on Monday December 01, 2025 @11:48AM (#46109625)
It's like saying "Star Wars is the best movie in the world" - to some people it is. To others it's terrible.

I feel like it would be advantageous to move away from a "one model fits all" mindset, and move towards a world where we have different genres of models that we use for different things.

The benchmark scores are turning into being just as useful as tomatometer movie scores. Something can score high, but if that's not the genre you like, the high score doesn't guarantee you'll like it.

[ Parent | on HN ]
Re:Google, Nvidia, and OpenAI – Stratechery by Ben Thompson by everdrive on Monday December 01, 2025 @11:58AM (#46109773)
Outside of experience and experimentation, is there a good way to know what models are strong for what tasks?
[ Parent | on HN ]
Re:Google, Nvidia, and OpenAI – Stratechery by Ben Thompson by grahamplace on Monday December 01, 2025 @12:11PM (#46109943)
See: https://lmarena.ai/leaderboard
[ Parent | on HN ]
Re:Google, Nvidia, and OpenAI – Stratechery by Ben Thompson by jasonjmcghee on Monday December 01, 2025 @12:18PM (#46110049)
Unless you overfit to benchmark style scenarios and are worse for real-world use.
[ Parent | on HN ]
Re:Google, Nvidia, and OpenAI – Stratechery by Ben Thompson by jpollock on Monday December 01, 2025 @12:29PM (#46110199)
Not really, it's like asking which C compiler was best back in the 90s.

You had Watcom, Intel, GCC, Borland, Microsoft, etc.

They all had different optimizations and different target markets.

Best to make your tooling model agnostic. I understand that tuned prompts are model _version_ specific, so you will need this anyways.

[ Parent | on HN ]