anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity

[Cherry-pick][v0.18.0.post1][PERF] Wan2.2 support adalayernorm on NPU (#2575)

Browse Source 
Signed-off-by: fan2956 <zhoufan53@huawei.com>
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
Co-authored-by: gcanlin <canlinguosdu@gmail.com>
This commit is contained in:
fan2956andgcanlin authored and GitHub committed 2026-04-16 08:47:38 +08:00
1 parent 91fa5c51d7
commit f70fd1f58b
4 files changed
+45 -81
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

vllm_omni/diffusion/cache/teacache/extractors.py
Vendored
+2 -1
View File
@@ -221,7 +221,8 @@ def extract_qwen_context(
block = module.transformer_blocks[0]
img_mod_params = block.img_mod(temb)
img_mod1, _ = img_mod_params.chunk(2, dim=-1)
img_modulated, _ = block.img_norm1(hidden_states, img_mod1)
img_scale1, img_shift1, _ = block._modulate(img_mod1)
img_modulated = block.img_norm1(hidden_states, img_scale1, img_shift1)
# ============================================================================
# DEFINE TRANSFORMER EXECUTION (Qwen-specific)