A LLaMA2-7b chatbot with memory running on CPU, and optimized using smooth quantization, 4-bit quantization or Intel® Extension For PyTorch with bfloat16.

A JAX implementation of stochastic addition.

A Pytorch implementation of stochastic addition.

A Python binding of Berkeley SoftFloat

End-to-end Vietnamese ASR pipeline using NVIDIA NeMo. Features production-grade CI/CD (GitHub Actions), hardware-aware optimization (40% latency reduction), and robust linguistic data testing.