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 14, 23:45 UTC
Handle non-serializable objects in vllm bench (#21665)
Huy Do committed
Jul 26, 2025 at 20:35 UTC
971948b846994819c50cb5c5f04a54822635e303
1 file changed
+6
-1
vllm/benchmarks/utils.py
+6
-1
@@ -67,4 +67,9 @@ class InfEncoder(json.JSONEncoder):
67
68
def write_to_json(filename: str, records: list) -> None:
69
with open(filename, "w") as f:
70
- json.dump(records, f, cls=InfEncoder)
70
+ json.dump(
71
+ records,
72
+ f,
73
+ cls=InfEncoder,
74
+ default=lambda o: f"<{type(o).__name__} is not JSON serializable>",
75
+ )