← Back to context Comment by johndough 4 days ago Why not? The model is 2.8T parameters with native MXFP4, which is 1400GB. 1 comment johndough Reply alightsoul 3 days ago I would figure it needs a lot of kV cache and context size which can be much bigger than the model itself
alightsoul 3 days ago I would figure it needs a lot of kV cache and context size which can be much bigger than the model itself
I would figure it needs a lot of kV cache and context size which can be much bigger than the model itself