OptiML Framework

OptiML is an advanced end-to-end framework designed for program synthesis and CUDA kernel optimization. Developed by researchers at the University of Central Florida, Florida International University, and Intel Corporation, it bridges the gap between functional code and high-performance hardware execution. The software uses a Mixture-of-Thoughts (MoT) generator to synthesize code and Monte Carlo Tree Search (MCTS) to refine it based on hardware profiler feedback.

The framework is a result of academic and industrial collaboration, primarily based in the United States. It is not a commercial product owned by a traditional consumer goods company but rather a specialized software tool within the machine learning and high-performance computing (HPC) ecosystem. Its development is tied to institutional research aimed at improving GPU performance for scientific computing.