Open-Source Artifacts

Visit our GitHub organization to explore all 24 repositories.

Loom

Python

An end-to-end compilation stack for mapping ML kernels onto spatial hardware architectures.

GitHub →

TileLoom

C++

An MLIR-backed compiler pipeline for exploring spatial hardware mappings and generating constraint models for dataflow accelerators.

GitHub →

Morpher v2

Python MIT

Open-source tool for CGRA accelerators — compilation, simulation, and design space exploration.

GitHub →

LISA

Python MIT

Portable framework for DFG mapping on spatial accelerators using graph neural networks.

GitHub →

FLEX

C++ MIT

CGRA with flexible execution paradigm supporting spatio-temporal vector dataflow.

GitHub →

DAOP

Python Apache 2.0

Data-Aware Offloading and Predictive Pre-Calculation for efficient MoE inference.

GitHub →

HyCUBE

C++

Simulator for the HyCUBE CGRA architecture with reconfigurable single-cycle multi-hop interconnect.

GitHub →

PANORAMA

C++

Divide-and-conquer approach for mapping complex loop kernels on CGRA.

GitHub →

Morpher

Python MIT

Original Morpher — integrated compilation and simulation framework for CGRA.

GitHub →

Chameleon

Dual memory replay for online continual learning on edge devices.

GitHub →