@@ -29,6 +29,10 @@ SVDQuant is a post-training quantization technique for 4-bit weights and activat
...
@@ -29,6 +29,10 @@ SVDQuant is a post-training quantization technique for 4-bit weights and activat
SVDQuant reduces the model size of the 12B FLUX.1 by 3.6×. Additionally, Nunchaku, further cuts memory usage of the 16-bit model by 3.5× and delivers 3.0× speedups over the NF4 W4A16 baseline on both the desktop and laptop NVIDIA RTX 4090 GPUs. Remarkably, on laptop 4090, it achieves in total 10.1× speedup by eliminating CPU offloading.
SVDQuant reduces the model size of the 12B FLUX.1 by 3.6×. Additionally, Nunchaku, further cuts memory usage of the 16-bit model by 3.5× and delivers 3.0× speedups over the NF4 W4A16 baseline on both the desktop and laptop NVIDIA RTX 4090 GPUs. Remarkably, on laptop 4090, it achieves in total 10.1× speedup by eliminating CPU offloading.
## Installation
## Installation
**Note**: We currently support only NVIDIA GPUs with architectures sm_86 (Ampere: RTX 3090, A6000), sm_89 (Ada: RTX 4090), and sm_80 (A100). See [this issue](https://github.com/mit-han-lab/nunchaku/issues/1) for more details.
