diff --git a/mobile/lib/services/machine_learning/face_ml/face_ml_service.dart b/mobile/lib/services/machine_learning/face_ml/face_ml_service.dart
index d88b8f041..6d3f8eec7 100644
--- a/mobile/lib/services/machine_learning/face_ml/face_ml_service.dart
+++ b/mobile/lib/services/machine_learning/face_ml/face_ml_service.dart
@@ -27,6 +27,7 @@ import "package:photos/models/file/extensions/file_props.dart";
 import "package:photos/models/file/file.dart";
 import "package:photos/models/file/file_type.dart";
 import "package:photos/models/ml/ml_versions.dart";
+import "package:photos/service_locator.dart";
 import 'package:photos/services/machine_learning/face_ml/face_clustering/face_clustering_service.dart';
 import "package:photos/services/machine_learning/face_ml/face_clustering/face_info_for_clustering.dart";
 import 'package:photos/services/machine_learning/face_ml/face_detection/detection.dart';
@@ -537,7 +538,8 @@ class FaceMlService {
             for (final f in chunk) {
               fileIds.add(f.uploadedFileID!);
             }
-            final EnteWatch? w = kDebugMode ? EnteWatch("face_em_fetch") : null;
+            final EnteWatch? w =
+                flagService.internalUser ? EnteWatch("face_em_fetch") : null;
             w?.start();
             w?.log('starting remote fetch for ${fileIds.length} files');
             final res =
@@ -566,6 +568,16 @@ class FaceMlService {
               remoteFileIdToVersion[fileMl.fileID] =
                   fileMl.faceEmbedding.version;
             }
+            if (res.noEmbeddingFileIDs.isNotEmpty) {
+              _logger.info(
+                'No embeddings found for ${res.noEmbeddingFileIDs.length} files',
+              );
+              for (final fileID in res.noEmbeddingFileIDs) {
+                faces.add(Face.empty(fileID, error: false));
+                remoteFileIdToVersion[fileID] = faceMlVersion;
+              }
+            }
+
             await FaceMLDataDB.instance.bulkInsertFaces(faces);
             w?.logAndReset('stored embeddings');
             for (final entry in remoteFileIdToVersion.entries) {
diff --git a/mobile/lib/services/machine_learning/file_ml/files_ml_data_response.dart b/mobile/lib/services/machine_learning/file_ml/files_ml_data_response.dart
index 665e89b7e..475f52d0a 100644
--- a/mobile/lib/services/machine_learning/file_ml/files_ml_data_response.dart
+++ b/mobile/lib/services/machine_learning/file_ml/files_ml_data_response.dart
@@ -2,11 +2,18 @@ import 'package:photos/services/machine_learning/file_ml/file_ml.dart';
 
 class FilesMLDataResponse {
   final Map<int, FileMl> mlData;
-  final Set<int> notIndexedFileIds;
-  final Set<int> fetchErrorFileIds;
+  // fileIDs that were indexed but they don't contain any meaningful embeddings
+  // and hence should be discarded for re-indexing
+  final Set<int> noEmbeddingFileIDs;
+  // fetchErrorFileIDs are the fileIDs for whom we failed failed to fetch embeddings
+  // from the storage
+  final Set<int> fetchErrorFileIDs;
+  // pendingIndexFileIDs are the fileIDs that were never indexed
+  final Set<int> pendingIndexFileIDs;
   FilesMLDataResponse(
     this.mlData, {
-    required this.notIndexedFileIds,
-    required this.fetchErrorFileIds,
+    required this.noEmbeddingFileIDs,
+    required this.fetchErrorFileIDs,
+    required this.pendingIndexFileIDs,
   });
 }
diff --git a/mobile/lib/services/machine_learning/file_ml/remote_fileml_service.dart b/mobile/lib/services/machine_learning/file_ml/remote_fileml_service.dart
index 95db6c381..3a8e383be 100644
--- a/mobile/lib/services/machine_learning/file_ml/remote_fileml_service.dart
+++ b/mobile/lib/services/machine_learning/file_ml/remote_fileml_service.dart
@@ -72,7 +72,8 @@ class RemoteFileMLService {
         },
       );
       final remoteEmb = res.data['embeddings'] as List;
-      final noFileIds = res.data['noDataFileIDs'] as List;
+      final pendingIndexFiles = res.data['pendingIndexFileIDs'] as List;
+      final noEmbeddingFiles = res.data['noEmbeddingFileIDs'] as List;
       final errFileIds = res.data['errFileIDs'] as List;
 
       final List<RemoteEmbedding> remoteEmbeddings = <RemoteEmbedding>[];
@@ -81,13 +82,14 @@ class RemoteFileMLService {
         remoteEmbeddings.add(embedding);
       }
 
-      final notIndexedFileIds = Set<int>.from(noFileIds.map((x) => x as int));
-      final fetchErrorFileIds = Set<int>.from(errFileIds.map((x) => x as int));
       final fileIDToFileMl = await decryptFileMLData(remoteEmbeddings);
       return FilesMLDataResponse(
         fileIDToFileMl,
-        notIndexedFileIds: notIndexedFileIds,
-        fetchErrorFileIds: fetchErrorFileIds,
+        noEmbeddingFileIDs:
+            Set<int>.from(noEmbeddingFiles.map((x) => x as int)),
+        fetchErrorFileIDs: Set<int>.from(errFileIds.map((x) => x as int)),
+        pendingIndexFileIDs:
+            Set<int>.from(pendingIndexFiles.map((x) => x as int)),
       );
     } catch (e, s) {
       _logger.severe("Failed to get embeddings", e, s);