Files
zeekayandHanzo Dev 6a363f7ee3 merge: upstream main ddf1b879 and the aarch64 repack branch 35d7ae7c
Three-way per file through the rename: base 39355c6c91, theirs the upstream
merge of main with 35d7ae7c (via 7ef6a694d), ours origin/main.

Lands the barrier thread pool, runtime CPU feature dispatch, aarch64 and x86
repacked quantized kernels (indexed_gemv, gemv_fused_shared_lhs), quantized
embedding on CPU/Metal/CUDA, paged flash-attn kernels, cutile fused MoE, the
in-place op aliasing API, parking_lot storage locks, Metal buffer builders and
debug labels, safetensors 0.8, hf-hub 1.0, tokenizers 0.23.

Ours keeps ROCm, Vulkan, wgpu, the IQ/MXFP4/NVFP4/TQ dtypes, and the i-quant
weight/activation block split in matmul. Our backends ride InplaceOpN and the
quantized embedding; Metal get_rows gains the IQ kernels. ug leaves, as upstream.

Co-authored-by: Hanzo Dev <[email protected]>
2026-09-15 15:52:28 -07:00
..
2026-01-16 12:30:55 -08:00

hiera

Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles This hanzo implementation uses pre-trained Hiera models from timm for inference. The classification head has been trained on the ImageNet dataset and returns the probabilities for the top-5 classes.

Running an example

$ cargo run --example  hiera --release  -- --image hanzo-ml-examples/examples/yolo-v8/assets/bike.jpg --which tiny
loaded image Tensor[dims 3, 224, 224; f32]
model built
mountain bike, all-terrain bike, off-roader: 71.15%
unicycle, monocycle     : 7.11%
knee pad                : 4.26%
crash helmet            : 1.48%
moped                   : 1.07%