diff --git a/pkg/scheduler/schedule_one.go b/pkg/scheduler/schedule_one.go
index bc6da0c1e32..a49861f7508 100644
--- a/pkg/scheduler/schedule_one.go
+++ b/pkg/scheduler/schedule_one.go
@@ -41,6 +41,7 @@ import (
 	"k8s.io/kubernetes/pkg/features"
 	"k8s.io/kubernetes/pkg/scheduler/framework"
 	"k8s.io/kubernetes/pkg/scheduler/framework/parallelize"
+	"k8s.io/kubernetes/pkg/scheduler/framework/plugins/dynamicresources"
 	"k8s.io/kubernetes/pkg/scheduler/metrics"
 	"k8s.io/kubernetes/pkg/scheduler/util"
 	utiltrace "k8s.io/utils/trace"
@@ -1109,6 +1110,16 @@ func (sched *Scheduler) assume(logger klog.Logger, state fwk.CycleState, assumed
 	// If the binding fails, scheduler will release resources allocated to assumed pod
 	// immediately.
 	assumedPodInfo.Pod.Spec.NodeName = host
+	if utilfeature.DefaultFeatureGate.Enabled(features.DRANodeAllocatableResources) {
+		// If DRANodeAllocatableResources is enabled, copy the calculated node allocatable resource claim status
+		// from the cycle state to the assumed pod's status. This ensures that the scheduler's
+		// cached version of the pod reflects the node allocatable resources allocated by the DRA plugin
+		// for this scheduling cycle, making this information available for NodeInfo cache update.
+		// Any potential NodeAllocatableResourceClaimStatuses from a previously failed scheduling attempt is overwritten.
+		// This field is not explicitly cleared as the Pod object is reconstructed in handleSchedulingFailure()
+		// before re-queueing.
+		assumedPodInfo.Pod.Status.NodeAllocatableResourceClaimStatuses = dynamicresources.ExtractPodNodeAllocatableResourceClaimStatus(logger, state, host)
+	}
 
 	if state.IsPodGroupSchedulingCycle() {
 		err := sched.nodeInfoSnapshot.AssumePod(assumedPodInfo.PodInfo)