onnxruntime
35b42a80 - Refactor Intel SubgroupMatrix MatMulNBits (#27911)

Commit
3 days ago
Refactor Intel SubgroupMatrix MatMulNBits (#27911) - Migrate inline C++ shader to WGSL templates - Add bias and weight index support for gpt_oss_20b - Enable xe-3lpg config for PTL
Author
Parents
Loading