@thumbsup.eth
Of the models in Venice, what's a better model for web-connected basic research. Stuff like comparison of similar things in the same category (political parties platforms, different phones/computers specifications, holdings in ETFs)?
Currently I'm alternating between DeepSeek R1 671B and Llama 3.1 405B, but I just noticed they've added Qwen 2.5 VL 72B.
I don't use AI a lot, and I'm not currently using it for any coding, but I've found it can save me time for this kind of trivial data hunting and comparison. That said, I don't really know how y'all decide which ones to use. Do you just try them all on every question?