← Back to context Comment by orbifold 3 days ago How did you setup the 2x dgx spark? 1 comment orbifold Reply colingauvin 3 days ago Pretty much just this, but tweaked to have I think 64 active conversations be cacheable, which drops the context pool to ~2M.https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Spark...I also had to downclock my Sparks (Asus) slightly, to 2000 Mhz. Otherwise they'd overheat and crash.
colingauvin 3 days ago Pretty much just this, but tweaked to have I think 64 active conversations be cacheable, which drops the context pool to ~2M.https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Spark...I also had to downclock my Sparks (Asus) slightly, to 2000 Mhz. Otherwise they'd overheat and crash.
Pretty much just this, but tweaked to have I think 64 active conversations be cacheable, which drops the context pool to ~2M.
https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Spark...
I also had to downclock my Sparks (Asus) slightly, to 2000 Mhz. Otherwise they'd overheat and crash.