Skip to content

[GDN] [Backport] optimize: avoid fla l2-norm recompilation by token count (#31558) - #27

Merged
apinge merged 2 commits into
apinge:qwen3_5_v0.5.15_dflashfrom
sammysun0711:cherry-pick/gdn-l2norm-dynamic-token-count
Aug 31, 2026
Merged

[GDN] [Backport] optimize: avoid fla l2-norm recompilation by token count (#31558)#27
apinge merged 2 commits into
apinge:qwen3_5_v0.5.15_dflashfrom
sammysun0711:cherry-pick/gdn-l2norm-dynamic-token-count

perf(gdn): fuse Q/K L2 normalization on AMD

c0f8a27
Select commit
Loading
Failed to load commit list.
DCO / DCO succeeded Aug 29, 2026 in 1s

DCO

All commits are signed off!

View more details on DCO