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 15, 09:45 UTC
[Misc] Call `ndarray.tobytes()` directly instead of `ndarray.data.tobytes()` (#18347)
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
Lukas Geiger committed
May 22, 2025 at 17:00 UTC
54631f826233dbd1c046f9a70e98bc2e25edff1a
1 file changed
+1
-1
vllm/multimodal/hasher.py
+1
-1
@@ -43,7 +43,7 @@ class MultiModalHasher:
43
"ndarray", {
44
"dtype": obj.dtype.str,
45
"shape": obj.shape,
46
- "data": obj.data.tobytes(),
46
+ "data": obj.tobytes(),
47
})
48
49
logger.warning(