A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 6, 12:00 UTC
[Bugfix] Fix ROCm crash in qwen3_next multi-stream events (#36795) (#37427)
Signed-off-by: JartX <sagformas@epdcenter.es>
JartX committed
Mar 18, 2026 at 21:06 UTC
a913b612d8a85a926c50815adb969056f10b62e2
1 file changed
+1
-1
vllm/model_executor/models/qwen3_next.py
+1
-1
@@ -427,7 +427,7 @@ class Qwen3NextGatedDeltaNet(nn.Module, MambaBase):
427
self.aux_stream = aux_stream()
428
self.events = (
429
[torch.cuda.Event(), torch.cuda.Event()]
430
- if current_platform.is_cuda()
430
+ if current_platform.is_cuda_alike()
431
else [None, None]
432
)
433