auto-round
f7898197
- feat: add W4A8 MoE Python API and perf/accuracy benchmark script
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
8 days ago
feat: add W4A8 MoE Python API and perf/accuracy benchmark script Co-authored-by: a32543254 <53296245+a32543254@users.noreply.github.com>
References
#2143 - feat: W4A8 ARK XPU MoE kernel (int4 weight / int8 compute) with prefill + decode
Author
Copilot
Parents
ef97754e
Loading