Hello! Is anyone else trying local AI? Im dont know what im doing but lately ive been trying "Llama 3,2 3b instruct q5km" on my phone and its been surprisingly good.
@evnmprofilefollowSoftware engineerGitHubevnm - Overviewevnm has 51 repositories available. Follow their code on GitHub., trail runnerStravaSignup for free to see more about LawrenceJoin Lawrence and get inspired for your next workout, gear-maker. Working on something new for homeownersgetonmain.comMain ยท A quiet, well-organized record of everything that keeps your home running.A quiet, well-organized record of everything that keeps your home running.replied to@candlewaxscribe๐profilefollowNew here, wanted to try textlog:
If you have Mac, you might check out this app: blaizzy.github.io/nativ/
@jamesjan3profilefollowresearcher, developer, dancerreplied to@candlewaxscribe๐profilefollowNew here, wanted to try textlog:
i think the best you can get right now is qwen 3.8 27B. youre gonna need 24 gigs of some fast memory though, i heard theres decent quantizations that can also make it run on 16. im quite excited about open/local models, but theres big investment overhead with the prices right now
@keitachikawa๐profilefollowaeres-u99 on github, CodeAtlas \o/replied to@candlewaxscribe๐profilefollowNew here, wanted to try textlog:
You might be able to try out colibri, its quite good and the models are a bit better. I am unable to connect any of the cli agents to do things with it though, probably due to context limits, but try it out. OlMoe is the lightest