… usability, and compatibility across the CUDA and DirectML builds. What’s New Live 3D Overhaul Live 3D is more responsive and stable, with reduced waviness, improved scen

…tion and extras for CPU, CUDA/TensorRT, OpenVINO, DirectML, QNN, CoreML, and other ONNX Runtime builds - MIT-licensed packaged CLIP tokenizer with no forced CPU runtime d

…tic hashes when the effective artifact changes. - DirectML session safety : Pinned ONNX Runtime 1.24.3, enforced DirectML-compatible session options, and serialized infer

…ds. - CoreML selection on Apple Silicon. - Opt-in DirectML, OpenVINO, TensorRT, ROCm, MIGraphX, WebGPU, CANN, QNN, and Vitis AI providers. - ANYLABELING DEVICE overrides,

…ixed; output matches the reference again. RIFE on DirectML is ~2x faster. Part of the model ran on the CPU. AMD and Intel GPUs now run all of it. --dynamic scale was wors

…on about why CPU fallback occured. Dependencies - DirectML version bumped to 1.8.2. - Microsoft.AI.MachineLearning version bumped to 1.11. - OnnxRuntime version bumped to