onnxruntime
97cfc86f
- Update GatherBlockQuantized to support 2-bits (#28530)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
67 days ago
Update GatherBlockQuantized to support 2-bits (#28530) ### Description Update GatherBlockQuantized to support 2-bits. Updated op schema, implemented the CPU and WebGPU EP. This helps to make the model smaller.
References
#28530 - Update GatherBlockQuantized to support 2-bits
Author
HectorSVC
Parents
7a3f9a21
Loading