We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 40a0b54 commit dce769eCopy full SHA for dce769e
vllm/worker/model_runner.py
@@ -46,7 +46,7 @@ def __init__(
46
47
self.graph_runners: Dict[int, CUDAGraphRunner] = {}
48
self.graph_memory_pool = None # Set during graph capture.
49
- self.device = self.model_config.device
+ self.device = self.model_config.device if model_config is not None else "cuda"
50
51
self.max_context_len_to_capture = (
52
self.model_config.max_context_len_to_capture
0 commit comments