#inference

zse-engine added to PyPI

Zero-dependency LLM inference engine. Fastest cold start, smallest VRAM footprint. CUDA + ROCm + Metal.