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 1, 21:15 UTC
[Bugfix] Fix gemma4 crash on CPU: guard mem_get_info call (#44615)
Signed-off-by: ADHITHYA BALAKRISHNAN <adhithya.balakrishnan@multicorewareinc.com>
adhithyamulticoreware committed
Jun 5, 2026 at 18:17 UTC
bbb6c274c80b58551ca3cabee674ddd66e1cfbab
1 file changed
+5
vllm/platforms/cpu.py
+5
@@ -92,6 +92,11 @@ class CpuPlatform(Platform):
92
93
return meminfo.total_memory
94
95
+ @classmethod
96
+ def mem_get_info(cls) -> tuple[int, int]:
97
+ meminfo = get_memory_node_info()
98
+ return meminfo.available_memory, meminfo.total_memory
99
+
100
@classmethod
101
def set_device(cls, device: torch.device) -> None:
102
"""