Introduction of coreml-stable-diffusion-mixed-bit-palettization
Model Details of coreml-stable-diffusion-mixed-bit-palettization
Core ML Stable Diffusion Mixed-Bit Palettization Resources
This repository contains resources related to the use of
mixed-bit palettization techniques for Stable Diffusion
. As explained in the original repo, this is an advanced compression technique that applies palettization using a varying number of bits per layer to keep quality above the desired threshold (as measured by the PSNR with respect to the original model). This can achieve strong space savings without resorting to learned quantization tables during fine-tuning.
You can apply any of the available recipes on the appropriate model to experiment and compare results.
unet-mbp-sdxl-1-base/
. Three pre-applied recipes on the UNet of Stable Diffusion XL 1.0 base:
Recipe
Effective bits
UNet size (GB)
Size reduction
recipe_3_41_bit_mixedpalette
3.41
1.1
77%
recipe_4_50_bit_mixedpalette
4.50
1.4
71%
recipe_6_55_bit_mixedpalette
6.55
2.0
58%
coreml-stable-diffusion-xl-base_mbp_4_50_palettized
. A full pipeline of a Core ML version of Stable Diffusion XL 1.0 base, with the
recipe_4_50_bit_mixedpalette
recipe pre-applied to the UNet. You can download it to use in your own Core ML apps.
This repository was prepared by Apple and Hugging Face in July 2023, from experiments conducted using public beta versions of iOS 17.0, iPadOS 17.0 and macOS 14.0.
Runs of apple coreml-stable-diffusion-mixed-bit-palettization on huggingface.co
74
Total runs
0
24-hour runs
9
3-day runs
17
7-day runs
73
30-day runs
More Information About coreml-stable-diffusion-mixed-bit-palettization huggingface.co Model
More coreml-stable-diffusion-mixed-bit-palettization license Visit here:
coreml-stable-diffusion-mixed-bit-palettization huggingface.co is an AI model on huggingface.co that provides coreml-stable-diffusion-mixed-bit-palettization's model effect (), which can be used instantly with this apple coreml-stable-diffusion-mixed-bit-palettization model. huggingface.co supports a free trial of the coreml-stable-diffusion-mixed-bit-palettization model, and also provides paid use of the coreml-stable-diffusion-mixed-bit-palettization. Support call coreml-stable-diffusion-mixed-bit-palettization model through api, including Node.js, Python, http.
apple coreml-stable-diffusion-mixed-bit-palettization online free
coreml-stable-diffusion-mixed-bit-palettization huggingface.co is an online trial and call api platform, which integrates coreml-stable-diffusion-mixed-bit-palettization's modeling effects, including api services, and provides a free online trial of coreml-stable-diffusion-mixed-bit-palettization, you can try coreml-stable-diffusion-mixed-bit-palettization online for free by clicking the link below.
apple coreml-stable-diffusion-mixed-bit-palettization online free url in huggingface.co:
coreml-stable-diffusion-mixed-bit-palettization is an open source model from GitHub that offers a free installation service, and any user can find coreml-stable-diffusion-mixed-bit-palettization on GitHub to install. At the same time, huggingface.co provides the effect of coreml-stable-diffusion-mixed-bit-palettization install, users can directly use coreml-stable-diffusion-mixed-bit-palettization installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
coreml-stable-diffusion-mixed-bit-palettization install url in huggingface.co: