Adding cuda kernel (optimized for sm80) for block-wise 4b quantized float 16 GEMM. #18619
Azure Pipelines / Linux OpenVINO CI Pipeline
succeeded
Feb 29, 2024 in 48m 6s
Build #20240228.28 succeeded
Details
- Failed: 0 (0.00%)
- Passed: 19,300 (99.88%)
- Other: 24 (0.12%)
- Total: 19,324
Loading