auto-round
07aa43e4
- fix: add patch for FP8 dequantization to handle ragged tiles in transformers
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
4 days ago
fix: add patch for FP8 dequantization to handle ragged tiles in transformers Signed-off-by: Xin He <xin3.he@intel.com>
Author
xin3he
Parents
5ea4e1ff
Loading