Heterogeneous Computing Python Packages

pmcxcl

Monte Carlo eXtreme for OpenCL (MCXCL)

gafime

GPU-Accelerated Feature Interaction Mining Engine (GAFIME) — native heterogeneous execution across Rust CPU/SIMD, CUDA, ROCm/HIP, and Metal.

laya-apple

Correctness-validated heterogeneous Laya runtime for Apple Silicon (MLX GPU + Apple Neural Engine)

tensortorrent

PyTorch compiler and Rust runtime for multi-CPU/GPU inference, NUMA-aware scheduling, model serving, parameter streaming, and activation spill.

qfabric

Research runtime for contract-driven recovery across Linux and RTOS execution domains on the Arduino UNO Q.

llnl-thicket

Toolkit for exploratory data analysis of ensemble performance data

flockparser

Distributed document RAG with VRAM-aware workload placement across heterogeneous nodes. Four interfaces: CLI, API, MCP, Web UI.

distpipe

A distributed framework to implement device-cloud collaborative workflow