Tune OV for latency

This commit is contained in:
Nicolas Mowen 2025-09-21 15:22:33 -06:00
parent cd519ed1ad
commit a155e1dc9d

View File

@ -194,6 +194,9 @@ class OpenVINOModelRunner(BaseModelRunner):
# Apply performance optimization
self.ov_core.set_property(device, {"PERF_COUNT": "NO"})
if device in ["GPU", "AUTO"]:
self.ov_core.set_property(device, {"PERFORMANCE_HINT": "LATENCY"})
# Compile model
self.compiled_model = self.ov_core.compile_model(
model=model_path, device_name=device