[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

![opdroid1234 Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::1096860280225742848.png) opdroid1234 [@opdroid1234](/creator/twitter/opdroid1234) on x 2171 followers
Created: 2025-07-24 22:17:27 UTC

i am usually in the build my own stuff camp, but I have been pretty happy with my choice of just jusing models via an api. its very unlikely that you are going to get the same tokens per second from your local rig that you do with the hosted llm api providers. So my recommednation would be macbook air with XX gig for client, ~60 bucks a month for a 24/7 dedicated ryzen hetzner box with XX gigs of ram and then XX bucks a month for whichever llm provider strikes your fancy.


XXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1948507849723670866/c:line.svg)

**Related Topics**
[gig](/topic/gig)
[llm](/topic/llm)

[Post Link](https://x.com/opdroid1234/status/1948507849723670866)