mirror of
https://github.com/hanzoai/ml.git
synced 2026-11-20 23:08:29 +00:00
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]>
hanzo-ml-mamba-minimal: minimal implementation of Mamba
This is based on mamba-minimal.
Compared to the mamba example, this version can handle training but is much slower.
Running the example
$ cargo run --example mamba-minimal --release -- --prompt "Mamba is the"
Mamba is the most popular and best-selling game in the world. It has been downloaded more than 1,000 times by over 1 million people worldwide since its release on March 18th 2016.
The Mamba series of games are a collection that combines elements from all genres including action, adventure, strategy & puzzle games with some unique gameplay features such as stealth and survival. The game is also known for its innovative graphics and the ability to play in a variety of different modes like single player or multiplayer.