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 4, 05:00 UTC
[Bugfix] Fix OOM in tensorizer LoRA deserialization (#41845)
Signed-off-by: Or Ozeri <oro@il.ibm.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Or Ozeri committed
May 7, 2026 at 12:17 UTC
6e6d182d18dbe5c81c9c9058a5571241f649ae7b
1 file changed
+1
vllm/lora/lora_model.py
+1
@@ -203,6 +203,7 @@ class LoRAModel:
203
tensors = TensorDeserializer(
204
lora_tensor_path,
205
dtype=tensorizer_config.dtype,
206
+ device=device,
207
**tensorizer_args.deserialization_kwargs,
208
)
209
check_unexpected_modules(tensors)