Commit Graph
3 Commits
Author SHA1 Message Date
Xinyi SongandThomas Wang c034120cb8 [AMD] Enable draft-extend CUDA graph and reduce bubble for MTP (#29202)
Co-authored-by: Thomas Wang <thomawan@amd.com>
2026-08-13 00:50:19 -07:00
Xinyi Song 1478cdec9f [AMD] Fuse Kimi-K3 attn-residual aggregation (#33599)
HIP Gated changes
2026-08-04 23:20:04 -07:00
Xinyi Song 371b96e210 [AMD] Fix DeepSeek-V4-Flash-FP8 on MI300 (#27972) 2026-06-12 01:51:32 -07:00