Property InferenceScheduling
InferenceScheduling
How the runtime allocates inference state and schedules requests onto it. Selected once for the process; the two strategies are never mixed and there is no fallback between them.
public static InferenceScheduling InferenceScheduling { get; set; }
Property Value
- InferenceScheduling
Default is PerRequestContext, the long-standing behaviour.
Remarks
Changing this after a model has begun serving requests does not migrate work already in flight; it governs what is built next.