Hugging Face
Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
HelloSun
/
sdcpp_zimage_opt
Like
0
GGUF
stable-diffusion.cpp
stable-diffusion
z-image
z-image-turbo
cpu
benchmark
amx
avx512
operator-fusion
License:
mit
Model card
Files
Files and versions
xet
Community
Copy to bucket
main
sdcpp_zimage_opt
/
patches
32.9 kB
Ctrl+K
Ctrl+K
1 contributor
History:
4 commits
HelloSun
VAE stage 3: vectorize direct-conv im2col (VAE decode ~4.7s -> ~3.9s, pixel-identical)
26151cf
1 day ago
ggml-0001-cpu-op-profile.patch
5.04 kB
Pipeline stage P1: shape-level profile diagnostic, flash-attn K packing once per head, build.sh submodule reset fix
1 day ago
ggml-0002-flash-attn-kpack.patch
8.21 kB
Pipeline stage P2: explain and fix refiner Q8_0 slowdown, V-pack hoist + 128x128 flash-attn tiles
1 day ago
ggml-0003-flash-attn-tiles.patch
355 Bytes
Pipeline stage P2: explain and fix refiner Q8_0 slowdown, V-pack hoist + 128x128 flash-attn tiles
1 day ago
ggml-0004-conv2d-im2col.patch
8.23 kB
VAE stage 3: vectorize direct-conv im2col (VAE decode ~4.7s -> ~3.9s, pixel-identical)
1 day ago
sdcpp-0001-zimage-ffn-w13-and-cpu-extra-buft.patch
11 kB
Add build/bench scripts, patches, benchmark results and model card
1 day ago
