Adding cuda kernel (optimized for sm80) for block-wise 4b quantized float 16 GEMM. #18619
Azure Pipelines / ONNX Runtime Web CI Pipeline (Build_wasm_Release build_WASM)
succeeded
Feb 29, 2024 in 55m 31s
Build_wasm_Release build_WASM succeeded
Loading