Why is it fixated on the front perspective? Interesting choice though, because most humans (and seems like other LLMs too) would pick a side perspective
As a lifelong windows (upto 10) and linux user, no I did not find MacOS (using as the primary os since 7 months) incredible in any sense of word in comparison. Only thing I like is the mac hardware
No only that, but per capita emissions of developed countries still remains higher. For example I found that US/Russia have 6x per capita emissions compared to India
I built my own slide rule in school for fun! It looked pretty cool to me at the time. The template is still out there if you search something like "paper slide rule".
With ollama you could offload a few layers to cpu if they don't fit in the VRAM. This will cost some performance ofcourse but it's much better than the alternative (everything on cpu)
Arxiv has about 2.6M articles, assuming about 10 pages per article, that's 26M pages. According to OpenAI, their cheapest embedding model (text-embedding-3-small) costs a dollar for 62.5K pages. So the price for calculating embedding for the whole Arxiv is about $416.
I think doing it locally with an open source model would be a lot cheaper as well. Especially because they wouldn't have to keep using OpenAI's API for each new query.
Edit: I overlooked the about page (https://searchthearxiv.com/about), seems like they *are* using OpenAI's API, but they only have 300K papers indexed, use an older embedding model, and only calculate embeddings on the abstract. So this should be pretty cheap.