onnxruntime
11e2f5c1 - [WebGPU] implements online softmax for non-flash attention webgpu ep (#29724)

Commit
31 days ago
[WebGPU] implements online softmax for non-flash attention webgpu ep (#29724) Builds on feature request from https://github.com/microsoft/onnxruntime/issues/29393 and online softmax implementation from https://github.com/microsoft/onnxruntime/pull/29694 --------- Co-authored-by: Prathik Rao <prathikrao@microsoft.com>
Author
Parents
Loading