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

[Hardware][Ascend] Accelerate VoxCPM2 LocDiT with NPUGraph (#7275)

Browse Source 
Signed-off-by: Big2Wheel <1404207590@qq.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
This commit is contained in:
yulunandHongsheng Liu authored and GitHub committed 2026-09-30 10:11:22 +08:00
1 parent 35cf6ac355
commit 9097fa6547
7 files changed
+277 -17
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

tests/model_executor/models/voxcpm2/test_runner_assisted_unified_graph_static.py
+2 -1
View File
@@ -192,9 +192,10 @@ def test_voxcpm2_scheduler_policy_stays_model_local():
assert "def _should_defer_waiting_admission(self) -> bool:" in common_source
assert "class VoxCPM2OmniARAsyncScheduler(OmniARAsyncScheduler)" in voxcpm2_scheduler_source
assert "import vllm_omni.platforms as omni_platform" in voxcpm2_scheduler_source
assert "from .runtime_config import _VoxCPM2RuntimeConfig" in voxcpm2_scheduler_source
assert "_VoxCPM2RuntimeConfig.from_vllm_config(self.vllm_config)" in voxcpm2_scheduler_source
assert "unified_decode_graph_available(use_cuda_graph=current_omni_platform.is_cuda())" in voxcpm2_scheduler_source
assert "use_cuda_graph=omni_platform.current_omni_platform.is_cuda()" in voxcpm2_scheduler_source
assert "_should_defer_waiting_for_unified_decode_graph" in voxcpm2_scheduler_source
assert "def schedule(" not in voxcpm2_scheduler_source
assert "create_request_queue" not in voxcpm2_scheduler_source