[GaggiX]

bitsandbytes is only used during training with these models tho (the 8-bit Adamw) quantizing the weights and the activations to a range of 256 values when the model needs to output a range 256 values creates noticeable artifacts as they are not going to map 1-to-1.

[liuliu]

Draw Things recently released a 8-bit quantized SD model that has comparable output as the FP16. It does use k-means based LUT and separate weights into blocks to minimize quantization errors.

[GaggiX]

I was going to search on the internet about it, but then I realized you are the author (and there is nothing online I think). I imagine that the activations are left in FP16 and the weights are converted in FP16 during inference, right?

Btw very cool

[liuliu]

Yes, computes are carried out in FP16 (so there is no compute efficiency gains, might be latency reductions due to memory-bandwidth saving). These savings are not realized yet because no custom kernels introduced yet.