#exolabs — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #exolabs, aggregated by home.social.
-
3천만원으로 윤리와 안전장치를 제공하지 않는 실용성 있는 Private AI 데이터 센터를 책상 위에 구축하기
-
3천만원으로 윤리와 안전장치를 제공하지 않는 실용성 있는 Private AI 데이터 센터를 책상 위에 구축하기
-
well sort of stark reality hitting with geerling vid on mini pc cluster ..//clustering servers with a couple 3090 looks like the goto spot for now until next gen parts in a year. the mini pc is great just not quite ready for prime time - they need to take these types of chips to the desktop and even server space, i think mem and io will jump in next gen #unified mem #apu #exolabs #llama
-
well sort of stark reality hitting with geerling vid on mini pc cluster ..//clustering servers with a couple 3090 looks like the goto spot for now until next gen parts in a year. the mini pc is great just not quite ready for prime time - they need to take these types of chips to the desktop and even server space, i think mem and io will jump in next gen #unified mem #apu #exolabs #llama
-
well sort of stark reality hitting with geerling vid on mini pc cluster ..//clustering servers with a couple 3090 looks like the goto spot for now until next gen parts in a year. the mini pc is great just not quite ready for prime time - they need to take these types of chips to the desktop and even server space, i think mem and io will jump in next gen #unified mem #apu #exolabs #llama
-
well sort of stark reality hitting with geerling vid on mini pc cluster ..//clustering servers with a couple 3090 looks like the goto spot for now until next gen parts in a year. the mini pc is great just not quite ready for prime time - they need to take these types of chips to the desktop and even server space, i think mem and io will jump in next gen #unified mem #apu #exolabs #llama
-
Cobbled together an #ExoLabs cluster to fuck around with #devstral a bit, since it's kinda too big for my M3 Max daily driver. While in the process of bringing up nodes the model hit a bug in the #MLX #Python module that deals with inference model sharding related to passing around MLX vs Numpy data structures.
For shits and giggles and also not being a top-tier #Numpy data structure debugging guy I asked Devstral to look at the bug and figure out a fix. After one wrong turn it came up with a fix which I applied to the other nodes and now it's happily sharding the bigger Devstral models. Not sure about vibe coding as a social contagion but from a “How close are we to #Skynet”-perspective I think we're cooked, chat.
Anyway enjoy your Memorial Day weekend 🎉
Figure 1. A very heterogeneous Exo cluster.
-
I built an AI supercomputer with 5 Mac Studios
- https://www.youtube.com/watch?v=Ju0ndy2kwlw -
deepseek v3 - big jump for chinese ai (hosted in china? be careful with creds) - it is open source to some degree - token cost is cheap but they need to be even more open and release smaller but even better models - the mkt wants the hw to run huge models like this locally through unified mem or cxl or on a very fast nvme #mac stack #exolabs #open weights
-
"
ETH: Präzise Schneemessung dank KI und SatellitenSo genau und schnell war die Schneemessung noch nie: ETH-Forschende haben eine künstliche Intelligenz entwickelt, die anhand von Satellitenbildern die Schneehöhe in der ganzen Schweiz bestimmen kann. Eine Pressemitteilung der ETH Zürich.
"
https://www.raumfahrer.net/eth-praezise-schneemessung-dank-ki-und-satelliten/14.12.2023
#AI #Copernicus #Erdbeobachtung #ESA #ETH #ExoLabs #GMES #KI #Raumfahrt #Satelliten #Schnee #Schweiz #Sentinel2