diff --git a/src/frontends/onnx/frontend/src/op/lstm.cpp b/src/frontends/onnx/frontend/src/op/lstm.cpp
index e004076c0e9ce6..e7bb52ea46e107 100644
--- a/src/frontends/onnx/frontend/src/op/lstm.cpp
+++ b/src/frontends/onnx/frontend/src/op/lstm.cpp
@@ -13,6 +13,8 @@
 #include "openvino/op/lstm_sequence.hpp"
 #include "openvino/op/multiply.hpp"
 #include "openvino/op/shape_of.hpp"
+#include "openvino/op/squeeze.hpp"
+#include "openvino/op/tile.hpp"
 #include "openvino/util/common_util.hpp"
 #include "utils/reshape.hpp"
 #include "utils/split.hpp"
@@ -37,6 +39,32 @@ enum class LSTMInput {
     LSTM_INPUT_P
 };
 
+// Helper function to reduce tensor rank to target_rank by squeezing leading dimensions.
+// Per ONNX specification, LSTM requires rank-3 inputs, so extra dimensions must be == 1.
+// If extra dimensions are != 1 at runtime, Squeeze will fail with a clear error.
+ov::Output<ov::Node> reduce_tensor_rank(const ov::Output<ov::Node>& input, int64_t target_rank) {
+    const auto& input_shape = input.get_partial_shape();
+
+    if (input_shape.rank().is_dynamic()) {
+        return input;
+    }
+
+    const auto input_rank = input_shape.rank().get_length();
+
+    if (input_rank <= target_rank) {
+        return input;
+    }
+
+    // Squeeze all leading dimensions to reduce rank to target_rank
+    std::vector<int64_t> axes_to_squeeze;
+    for (int64_t i = 0; i < input_rank - target_rank; ++i) {
+        axes_to_squeeze.push_back(i);
+    }
+
+    auto axes_const = v0::Constant::create(ov::element::i64, Shape{axes_to_squeeze.size()}, axes_to_squeeze);
+    return std::make_shared<v0::Squeeze>(input, axes_const);
+}
+
 struct LSTMNgInputMap {
     explicit LSTMNgInputMap(const Node& node) {
         const auto& ng_inputs = node.get_ov_inputs();
@@ -48,7 +76,14 @@ struct LSTMNgInputMap {
         // Packed input sequences.
         // ONNX Shape: [seq_length, batch_size, input_size]
         // OpenVino Shape: [batch_size, seq_length, input_size]
-        m_input_map[LSTMInput::LSTM_INPUT_X] = ov::op::util::reorder_axes(ng_inputs.at(0), {1, 0, 2});
+
+        // First reduce rank if needed, THEN reorder axes
+        // This is important because Squeeze changes dimension indices
+        auto input_x = ng_inputs.at(0);
+        input_x = reduce_tensor_rank(input_x, 3);
+        input_x = ov::op::util::reorder_axes(input_x, {1, 0, 2});
+
+        m_input_map[LSTMInput::LSTM_INPUT_X] = input_x;
 
         // Weight tensor for the gates.
         // Shape: [num_directions, 4*hidden_size, input_size]
@@ -124,7 +159,12 @@ struct LSTMNgInputMap {
         // ONNX Shape: [num_directions, batch_size, hidden_size]
         // OpenVino Shape: [batch_size, num_directions, hidden_size]
         if (ng_inputs.size() > 5 && !ov::op::util::is_null(ng_inputs.at(5))) {
-            m_input_map[LSTMInput::LSTM_INPUT_INIT_H] = ov::op::util::reorder_axes(ng_inputs.at(5), {1, 0, 2});
+            auto init_h = ng_inputs.at(5);
+            // First reduce rank, THEN reorder axes
+            init_h = reduce_tensor_rank(init_h, 3);
+            init_h = ov::op::util::reorder_axes(init_h, {1, 0, 2});
+
+            m_input_map[LSTMInput::LSTM_INPUT_INIT_H] = init_h;
         } else {
             auto init_h_shape =
                 std::make_shared<v0::Concat>(ov::OutputVector{batch_size_node, num_directions_node, hidden_size_node},
@@ -137,7 +177,12 @@ struct LSTMNgInputMap {
         // ONNX Shape: [num_directions, batch_size, hidden_size]
         // OpenVino Shape: [batch_size, num_directions, hidden_size]
         if (ng_inputs.size() > 6 && !ov::op::util::is_null(ng_inputs.at(6))) {
-            m_input_map[LSTMInput::LSTM_INPUT_INIT_C] = ov::op::util::reorder_axes(ng_inputs.at(6), {1, 0, 2});
+            auto init_c = ng_inputs.at(6);
+            // First reduce rank, THEN reorder axes
+            init_c = reduce_tensor_rank(init_c, 3);
+            init_c = ov::op::util::reorder_axes(init_c, {1, 0, 2});
+
+            m_input_map[LSTMInput::LSTM_INPUT_INIT_C] = init_c;
         } else {
             auto init_c_shape =
                 std::make_shared<v0::Concat>(ov::OutputVector{batch_size_node, num_directions_node, hidden_size_node},
diff --git a/src/frontends/onnx/tests/models/lstm_high_rank_input.prototxt b/src/frontends/onnx/tests/models/lstm_high_rank_input.prototxt
new file mode 100644
index 00000000000000..7ced96e260b29a
--- /dev/null
+++ b/src/frontends/onnx/tests/models/lstm_high_rank_input.prototxt
@@ -0,0 +1,167 @@
+ir_version: 4
+producer_name: "OpenVINO ONNX Frontend"
+graph {
+  node {
+    input: "X"
+    input: "W"
+    input: "R"
+    input: "B"
+    output: "Y"
+    output: "Y_h"
+    output: "Y_c"
+    op_type: "LSTM"
+    attribute {
+      name: "direction"
+      s: "forward"
+      type: STRING
+    }
+    attribute {
+      name: "hidden_size"
+      i: 2
+      type: INT
+    }
+  }
+  name: "compute_graph"
+  input {
+    name: "X"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 3
+          }
+          dim {
+            dim_value: 2
+          }
+          dim {
+            dim_value: 4
+          }
+        }
+      }
+    }
+  }
+  input {
+    name: "W"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 8
+          }
+          dim {
+            dim_value: 4
+          }
+        }
+      }
+    }
+  }
+  input {
+    name: "R"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 8
+          }
+          dim {
+            dim_value: 2
+          }
+        }
+      }
+    }
+  }
+  input {
+    name: "B"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 16
+          }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 3
+          }
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 2
+          }
+          dim {
+            dim_value: 2
+          }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y_h"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 2
+          }
+          dim {
+            dim_value: 2
+          }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y_c"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 2
+          }
+          dim {
+            dim_value: 2
+          }
+        }
+      }
+    }
+  }
+}
+opset_import {
+  version: 7
+}
diff --git a/src/frontends/onnx/tests/models/lstm_rank4_with_unsqueeze.prototxt b/src/frontends/onnx/tests/models/lstm_rank4_with_unsqueeze.prototxt
new file mode 100644
index 00000000000000..313dc2892b405b
--- /dev/null
+++ b/src/frontends/onnx/tests/models/lstm_rank4_with_unsqueeze.prototxt
@@ -0,0 +1,587 @@
+ir_version: 9
+producer_name: "OpenVINO Test"
+graph {
+  node {
+    input: "X_input"
+    input: "unsqueeze_axes"
+    output: "X_unsqueezed"
+    op_type: "Unsqueeze"
+  }
+  node {
+    input: "X_unsqueezed"
+    input: "W"
+    input: "R"
+    input: "B"
+    output: "Y"
+    output: "Y_h"
+    output: "Y_c"
+    op_type: "LSTM"
+    attribute {
+      name: "direction"
+      s: "forward"
+      type: STRING
+    }
+    attribute {
+      name: "hidden_size"
+      i: 8
+      type: INT
+    }
+  }
+  name: "lstm_rank4_with_unsqueeze"
+  initializer {
+    dims: 1
+    dims: 32
+    dims: 4
+    data_type: 1
+    float_data: 0.496714145
+    float_data: -0.138264298
+    float_data: 0.647688568
+    float_data: 1.5230298
+    float_data: -0.234153375
+    float_data: -0.234136954
+    float_data: 1.57921278
+    float_data: 0.767434716
+    float_data: -0.469474375
+    float_data: 0.542560041
+    float_data: -0.463417679
+    float_data: -0.465729743
+    float_data: 0.241962269
+    float_data: -1.91328025
+    float_data: -1.72491789
+    float_data: -0.562287509
+    float_data: -1.01283109
+    float_data: 0.31424734
+    float_data: -0.908024073
+    float_data: -1.41230369
+    float_data: 1.46564877
+    float_data: -0.2257763
+    float_data: 0.0675282
+    float_data: -1.42474818
+    float_data: -0.544382751
+    float_data: 0.11092259
+    float_data: -1.15099359
+    float_data: 0.37569803
+    float_data: -0.600638688
+    float_data: -0.291693747
+    float_data: -0.601706624
+    float_data: 1.85227823
+    float_data: -0.013497225
+    float_data: -1.05771089
+    float_data: 0.822544932
+    float_data: -1.22084367
+    float_data: 0.208863601
+    float_data: -1.95967007
+    float_data: -1.32818604
+    float_data: 0.196861237
+    float_data: 0.738466561
+    float_data: 0.171368286
+    float_data: -0.115648285
+    float_data: -0.301103681
+    float_data: -1.47852194
+    float_data: -0.719844222
+    float_data: -0.460638762
+    float_data: 1.05712223
+    float_data: 0.343618304
+    float_data: -1.76304018
+    float_data: 0.324083984
+    float_data: -0.385082275
+    float_data: -0.676922
+    float_data: 0.611676276
+    float_data: 1.03099954
+    float_data: 0.931280136
+    float_data: -0.839217544
+    float_data: -0.309212387
+    float_data: 0.331263423
+    float_data: 0.975545108
+    float_data: -0.479174227
+    float_data: -0.185658976
+    float_data: -1.10633492
+    float_data: -1.19620657
+    float_data: 0.812525809
+    float_data: 1.35624
+    float_data: -0.0720101222
+    float_data: 1.00353289
+    float_data: 0.361636
+    float_data: -0.645119727
+    float_data: 0.361395597
+    float_data: 1.53803658
+    float_data: -0.0358260386
+    float_data: 1.56464362
+    float_data: -2.61974502
+    float_data: 0.821902514
+    float_data: 0.0870470703
+    float_data: -0.299007356
+    float_data: 0.0917607769
+    float_data: -1.98756886
+    float_data: -0.21967189
+    float_data: 0.357112557
+    float_data: 1.47789407
+    float_data: -0.518270195
+    float_data: -0.808493614
+    float_data: -0.501757
+    float_data: 0.915402114
+    float_data: 0.328751117
+    float_data: -0.529760182
+    float_data: 0.513267457
+    float_data: 0.0970775485
+    float_data: 0.968645
+    float_data: -0.70205307
+    float_data: -0.32766214
+    float_data: -0.392108142
+    float_data: -1.46351492
+    float_data: 0.296120286
+    float_data: 0.261055261
+    float_data: 0.00511345686
+    float_data: -0.234587133
+    float_data: -1.4153707
+    float_data: -0.420645326
+    float_data: -0.342714518
+    float_data: -0.802277267
+    float_data: -0.161285713
+    float_data: 0.404050857
+    float_data: 1.88618588
+    float_data: 0.174577817
+    float_data: 0.257550389
+    float_data: -0.0744459182
+    float_data: -1.91877127
+    float_data: -0.0265138745
+    float_data: 0.0602302104
+    float_data: 2.46324205
+    float_data: -0.192360967
+    float_data: 0.301547348
+    float_data: -0.0347117707
+    float_data: -1.16867805
+    float_data: 1.14282286
+    float_data: 0.751933038
+    float_data: 0.791031957
+    float_data: -0.909387469
+    float_data: 1.40279436
+    float_data: -1.40185106
+    float_data: 0.58685708
+    float_data: 2.19045568
+    float_data: -0.990536332
+    float_data: -0.56629771
+    name: "W"
+  }
+  initializer {
+    dims: 1
+    dims: 32
+    dims: 8
+    data_type: 1
+    float_data: 0.0996513665
+    float_data: -0.503475666
+    float_data: -1.55066347
+    float_data: 0.068562977
+    float_data: -1.06230366
+    float_data: 0.47359243
+    float_data: -0.919424236
+    float_data: 1.54993439
+    float_data: -0.783253312
+    float_data: -0.322061509
+    float_data: 0.813517213
+    float_data: -1.23086429
+    float_data: 0.227459937
+    float_data: 1.30714273
+    float_data: -1.60748327
+    float_data: 0.184633866
+    float_data: 0.259882808
+    float_data: 0.78182286
+    float_data: -1.23695076
+    float_data: -1.32045662
+    float_data: 0.521941543
+    float_data: 0.296984673
+    float_data: 0.250492841
+    float_data: 0.346448213
+    float_data: -0.680024743
+    float_data: 0.2322537
+    float_data: 0.293072462
+    float_data: -0.714351416
+    float_data: 1.86577451
+    float_data: 0.473832935
+    float_data: -1.19130349
+    float_data: 0.656553626
+    float_data: -0.974681675
+    float_data: 0.787084579
+    float_data: 1.15859556
+    float_data: -0.820682347
+    float_data: 0.963376105
+    float_data: 0.412780941
+    float_data: 0.822060168
+    float_data: 1.89679301
+    float_data: -0.24538812
+    float_data: -0.753736138
+    float_data: -0.889514446
+    float_data: -0.815810263
+    float_data: -0.0771017075
+    float_data: 0.341151983
+    float_data: 0.276690811
+    float_data: 0.827183247
+    float_data: 0.0130018918
+    float_data: 1.45353413
+    float_data: -0.264656842
+    float_data: 2.72016907
+    float_data: 0.625667334
+    float_data: -0.857157528
+    float_data: -1.07089245
+    float_data: 0.48247242
+    float_data: -0.22346279
+    float_data: 0.714000523
+    float_data: 0.473237634
+    float_data: -0.0728289112
+    float_data: -0.846793711
+    float_data: -1.51484728
+    float_data: -0.446514964
+    float_data: 0.856398821
+    float_data: 0.214093745
+    float_data: -1.24573874
+    float_data: 0.173180923
+    float_data: 0.385317385
+    float_data: -0.883857429
+    float_data: 0.153725103
+    float_data: 0.0582087189
+    float_data: -1.14297032
+    float_data: 0.357787371
+    float_data: 0.560784519
+    float_data: 1.0830512
+    float_data: 1.05380201
+    float_data: -1.37766933
+    float_data: -0.937825
+    float_data: 0.515035272
+    float_data: 0.513785958
+    float_data: 0.515047669
+    float_data: 3.85273147
+    float_data: 0.570890486
+    float_data: 1.13556564
+    float_data: 0.954001784
+    float_data: 0.651391268
+    float_data: -0.315269232
+    float_data: 0.758969247
+    float_data: -0.772825241
+    float_data: -0.236818612
+    float_data: -0.485363543
+    float_data: 0.0818741396
+    float_data: 2.31465864
+    float_data: -1.86726522
+    float_data: 0.686260164
+    float_data: -1.61271584
+    float_data: -0.471931875
+    float_data: 1.08895063
+    float_data: 0.0642800182
+    float_data: -1.07774472
+    float_data: -0.715303719
+    float_data: 0.679597735
+    float_data: -0.730366647
+    float_data: 0.216458589
+    float_data: 0.0455718413
+    float_data: -0.651600361
+    float_data: 2.14394403
+    float_data: 0.633919
+    float_data: -2.02514267
+    float_data: 0.186454311
+    float_data: -0.661786437
+    float_data: 0.852433324
+    float_data: -0.792520761
+    float_data: -0.114736438
+    float_data: 0.504987299
+    float_data: 0.8657552
+    float_data: -1.2002964
+    float_data: -0.334501237
+    float_data: -0.474945307
+    float_data: -0.653329253
+    float_data: 1.76545429
+    float_data: 0.404981703
+    float_data: -1.26088393
+    float_data: 0.917861938
+    float_data: 2.12215614
+    float_data: 1.03246522
+    float_data: -1.51937
+    float_data: -0.484234065
+    float_data: 1.26691115
+    float_data: -0.707669437
+    float_data: 0.443819433
+    float_data: 0.774634063
+    float_data: -0.926930487
+    float_data: -0.0595253557
+    float_data: -3.24126744
+    float_data: -1.0243876
+    float_data: -0.252568156
+    float_data: -1.24778318
+    float_data: 1.63241136
+    float_data: -1.43014133
+    float_data: -0.440044492
+    float_data: 0.130740583
+    float_data: 1.44127333
+    float_data: -1.43586218
+    float_data: 1.16316378
+    float_data: 0.0102330614
+    float_data: -0.981508672
+    float_data: 0.462103486
+    float_data: 0.199059695
+    float_data: -0.600216866
+    float_data: 0.0698020831
+    float_data: -0.3853136
+    float_data: 0.113517344
+    float_data: 0.662130654
+    float_data: 1.58601677
+    float_data: -1.2378155
+    float_data: 2.13303328
+    float_data: -1.95208776
+    float_data: -0.151785091
+    float_data: 0.588317215
+    float_data: 0.280991882
+    float_data: -0.622699499
+    float_data: -0.208122253
+    float_data: -0.493000925
+    float_data: -0.589364767
+    float_data: 0.849602103
+    float_data: 0.357015491
+    float_data: -0.692909598
+    float_data: 0.89959985
+    float_data: 0.307299525
+    float_data: 0.812862098
+    float_data: 0.629628837
+    float_data: -0.828995
+    float_data: -0.560181
+    float_data: 0.747293591
+    float_data: 0.610370278
+    float_data: -0.0209015943
+    float_data: 0.117327385
+    float_data: 1.2776649
+    float_data: -0.591571391
+    float_data: 0.547097385
+    float_data: -0.202192649
+    float_data: -0.217681199
+    float_data: 1.09877682
+    float_data: 0.825416327
+    float_data: 0.813509643
+    float_data: 1.30547881
+    float_data: 0.0210038424
+    float_data: 0.681952953
+    float_data: -0.310266763
+    float_data: 0.324166358
+    float_data: -0.130143061
+    float_data: 0.0969959646
+    float_data: 0.595157
+    float_data: -0.818220675
+    float_data: 2.0923872
+    float_data: -1.00601733
+    float_data: -1.21418858
+    float_data: 1.15811086
+    float_data: 0.791662693
+    float_data: 0.624119818
+    float_data: 0.62834549
+    float_data: -0.0122467726
+    float_data: -0.897254348
+    float_data: 0.0758045614
+    float_data: -0.677161694
+    float_data: 0.97511971
+    float_data: -0.147057384
+    float_data: -0.82549721
+    float_data: -0.321385831
+    float_data: 0.412931442
+    float_data: -0.563724577
+    float_data: -0.822220385
+    float_data: 0.243687212
+    float_data: 0.244966567
+    float_data: -0.506943166
+    float_data: -0.471038312
+    float_data: 0.232049942
+    float_data: -1.44808435
+    float_data: -1.40746379
+    float_data: -0.718444228
+    float_data: -0.213447154
+    float_data: 0.310907573
+    float_data: 1.47535622
+    float_data: 0.857659638
+    float_data: -0.159938529
+    float_data: -0.0190162081
+    float_data: -1.00252938
+    float_data: -0.0185131356
+    float_data: -0.288658649
+    float_data: 0.322718561
+    float_data: -0.82723093
+    float_data: 0.519346535
+    float_data: 1.53273892
+    float_data: -0.108760148
+    float_data: 0.401711732
+    float_data: 0.690144
+    float_data: -0.401220471
+    float_data: 0.224092484
+    float_data: 0.0125924
+    float_data: 0.0976761
+    float_data: -0.773009777
+    float_data: 0.024510175
+    float_data: 0.497998297
+    float_data: 1.45114362
+    float_data: 0.959270835
+    float_data: 2.15318251
+    float_data: -0.767347574
+    float_data: 0.872320652
+    float_data: 0.18334201
+    float_data: 2.18980289
+    float_data: -0.80829829
+    float_data: -0.839721859
+    float_data: -0.599392653
+    float_data: -2.12389565
+    float_data: -0.525755048
+    name: "R"
+  }
+  initializer {
+    dims: 1
+    dims: 64
+    data_type: 1
+    float_data: -0.759132683
+    float_data: 0.150393784
+    float_data: 0.341756
+    float_data: 1.87617087
+    float_data: 0.950423837
+    float_data: -0.576903641
+    float_data: -0.898414671
+    float_data: 0.49191916
+    float_data: -1.32023323
+    float_data: 1.83145881
+    float_data: 1.17944014
+    float_data: -0.469175667
+    float_data: -1.71313453
+    float_data: 1.35387242
+    float_data: -0.114539847
+    float_data: 1.23781633
+    float_data: -1.5944277
+    float_data: -0.599375
+    float_data: 0.00524369953
+    float_data: 0.0469805934
+    float_data: -0.450065464
+    float_data: 0.622849941
+    float_data: -1.0676204
+    float_data: -0.142379478
+    float_data: 0.120295629
+    float_data: 0.514438808
+    float_data: 0.711614907
+    float_data: -1.12464213
+    float_data: -1.53411412
+    float_data: 1.27767682
+    float_data: 0.332314
+    float_data: -0.748486519
+    float_data: 1.55115199
+    float_data: 0.115674637
+    float_data: 1.17929721
+    float_data: 0.0675184801
+    float_data: 2.06074786
+    float_data: 1.75534081
+    float_data: -0.248964146
+    float_data: 0.971570969
+    float_data: 0.645375967
+    float_data: 1.3686316
+    float_data: -0.964923441
+    float_data: 0.686051488
+    float_data: 1.05842447
+    float_data: -1.75873947
+    float_data: -1.18325853
+    float_data: -2.03923225
+    float_data: -0.269406825
+    float_data: 0.717542231
+    float_data: 1.50235701
+    float_data: 0.0740947798
+    float_data: 1.6286155
+    float_data: -1.38010144
+    float_data: -1.70338249
+    float_data: -0.0555477
+    float_data: 0.384065449
+    float_data: -0.0326947495
+    float_data: -2.06744218
+    float_data: -0.0891200379
+    float_data: -1.30446947
+    float_data: 0.669672549
+    float_data: 0.366598248
+    float_data: -0.939879775
+    name: "B"
+  }
+  initializer {
+    dims: 1
+    data_type: 7
+    int64_data: 0
+    name: "unsqueeze_axes"
+  }
+  input {
+    name: "X_input"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 3
+          }
+          dim {
+            dim_value: 2
+          }
+          dim {
+            dim_value: 4
+          }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 3
+          }
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 2
+          }
+          dim {
+            dim_value: 8
+          }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y_h"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 2
+          }
+          dim {
+            dim_value: 8
+          }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y_c"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 2
+          }
+          dim {
+            dim_value: 8
+          }
+        }
+      }
+    }
+  }
+}
+opset_import {
+  version: 13
+}
diff --git a/src/frontends/onnx/tests/models/lstm_rank5_squeeze.prototxt b/src/frontends/onnx/tests/models/lstm_rank5_squeeze.prototxt
new file mode 100644
index 00000000000000..b45a56b0b67999
--- /dev/null
+++ b/src/frontends/onnx/tests/models/lstm_rank5_squeeze.prototxt
@@ -0,0 +1,122 @@
+ir_version: 7
+producer_name: "OpenVINO LSTM Test"
+graph {
+  node {
+    input: "X"
+    input: "W"
+    input: "R"
+    input: "B"
+    output: "Y"
+    output: "Y_h"
+    output: "Y_c"
+    op_type: "LSTM"
+    attribute {
+      name: "direction"
+      s: "forward"
+      type: STRING
+    }
+    attribute {
+      name: "hidden_size"
+      i: 2
+      type: INT
+    }
+  }
+  name: "lstm_rank5_squeeze"
+  input {
+    name: "X"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim { dim_value: 1 }
+          dim { dim_value: 1 }
+          dim { dim_value: 3 }
+          dim { dim_value: 2 }
+          dim { dim_value: 4 }
+        }
+      }
+    }
+  }
+  input {
+    name: "W"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim { dim_value: 1 }
+          dim { dim_value: 8 }
+          dim { dim_value: 4 }
+        }
+      }
+    }
+  }
+  input {
+    name: "R"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim { dim_value: 1 }
+          dim { dim_value: 8 }
+          dim { dim_value: 2 }
+        }
+      }
+    }
+  }
+  input {
+    name: "B"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim { dim_value: 1 }
+          dim { dim_value: 16 }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim { dim_value: 3 }
+          dim { dim_value: 1 }
+          dim { dim_value: 2 }
+          dim { dim_value: 2 }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y_h"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim { dim_value: 1 }
+          dim { dim_value: 2 }
+          dim { dim_value: 2 }
+        }
+      }
+    }
+  }
+  output {
+    name: "Y_c"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim { dim_value: 1 }
+          dim { dim_value: 2 }
+          dim { dim_value: 2 }
+        }
+      }
+    }
+  }
+}
+opset_import {
+  domain: ""
+  version: 14
+}
diff --git a/src/frontends/onnx/tests/onnx_import_rnn.in.cpp b/src/frontends/onnx/tests/onnx_import_rnn.in.cpp
index db26d0b0245c3b..3c333f0b443cc1 100644
--- a/src/frontends/onnx/tests/onnx_import_rnn.in.cpp
+++ b/src/frontends/onnx/tests/onnx_import_rnn.in.cpp
@@ -425,6 +425,190 @@ OPENVINO_TEST(${BACKEND_NAME}, onnx_model_lstm_dynamic_batch_size_and_seq_len) {
     test_case.run(DEFAULT_FLOAT_TOLERANCE_BITS + 1);
 }
 
+// Test for LSTM with high-rank input (rank reduction)
+// Tests the reduce_tensor_rank() function in lstm.cpp
+OPENVINO_TEST(${BACKEND_NAME}, onnx_model_lstm_rank5_squeeze) {
+    if (std::string("${BACKEND_NAME}") == std::string("IE_GPU")) {
+        GTEST_SKIP() << "GPU backend has known accuracy issues with LSTMSequence Y_h/Y_c outputs. "
+                     << "Root cause: fp16/fp32 buffer mismatch in OneDNN integration.";
+    }
+
+    auto model = convert_model("lstm_rank5_squeeze.onnx");
+    auto test_case = ov::test::TestCase(model, s_device);
+
+    // Input X shape: [1, 1, 3, 2, 4]
+    std::vector<float> X = {0.8125258088111877f,   1.3562400341033936f,  -0.07201012223958969f, 1.003532886505127f,
+                            0.3616360127925873f,   -0.6451197266578674f, 0.36139559745788574f,  1.538036584854126f,
+                            -0.03582603856921196f, 1.5646436214447021f,  -2.6197450160980225f,  0.8219025135040283f,
+                            0.08704707026481628f,  -0.2990073561668396f, 0.0917607769370079f,   -1.9875688552856445f,
+                            -0.21967189013957977f, 0.3571125566959381f,  1.4778940677642822f,   -0.5182701945304871f,
+                            -0.8084936141967773f,  -0.501757025718689f,  0.9154021143913269f,   0.3287511169910431f};
+
+    // Weight W shape: [1, 8, 4]
+    std::vector<float> W = {0.49671414494514465f, -0.13826429843902588f, 0.6476885676383972f,  1.5230298042297363f,
+                            -0.2341533750295639f, -0.23413695394992828f, 1.5792127847671509f,  0.7674347162246704f,
+                            -0.4694743752479553f, 0.5425600409507751f,   -0.4634176790714264f, -0.4657297432422638f,
+                            0.241962268948555f,   -1.9132802486419678f,  -1.7249178886413574f, -0.5622875094413757f,
+                            -1.0128310918807983f, 0.31424733996391296f,  -0.9080240726470947f, -1.4123036861419678f,
+                            1.4656487703323364f,  -0.2257762998342514f,  0.06752820312976837f, -1.424748182296753f,
+                            -0.5443827509880066f, 0.11092258989810944f,  -1.1509935855865479f, 0.3756980299949646f,
+                            -0.6006386876106262f, -0.2916937470436096f,  -0.6017066240310669f, 1.852278232574463f};
+
+    // Recurrent weight R shape: [1, 8, 2]
+    std::vector<float> R = {-0.013497225008904934f,
+                            -1.057710886001587f,
+                            0.8225449323654175f,
+                            -1.2208436727523804f,
+                            0.20886360108852386f,
+                            -1.959670066833496f,
+                            -1.32818603515625f,
+                            0.19686123728752136f,
+                            0.7384665608406067f,
+                            0.1713682860136032f,
+                            -0.1156482845544815f,
+                            -0.3011036813259125f,
+                            -1.4785219430923462f,
+                            -0.7198442220687866f,
+                            -0.46063876152038574f,
+                            1.0571222305297852f};
+
+    // Bias B shape: [1, 16]
+    std::vector<float> B = {0.3436183035373688f,
+                            -1.7630401849746704f,
+                            0.32408398389816284f,
+                            -0.38508227467536926f,
+                            -0.6769220232963562f,
+                            0.6116762757301331f,
+                            1.0309995412826538f,
+                            0.9312801361083984f,
+                            -0.8392175436019897f,
+                            -0.3092123866081238f,
+                            0.3312634229660034f,
+                            0.9755451083183289f,
+                            -0.4791742265224457f,
+                            -0.18565897643566132f,
+                            -1.106334924697876f,
+                            -1.1962065696716309f};
+
+    // Expected outputs from ONNX Runtime
+    // Y shape: [3, 1, 2, 2]
+    std::vector<float> expected_Y = {0.04541102051734924f,
+                                     0.007742831949144602f,
+                                     -0.05228624865412712f,
+                                     0.24427476525306702f,
+                                     0.24919648468494415f,
+                                     0.016252165660262108f,
+                                     -0.10558786988258362f,
+                                     0.337483674287796f,
+                                     -0.1856241226196289f,
+                                     -0.027182353660464287f,
+                                     -0.1143757626414299f,
+                                     0.14773108065128326f};
+
+    // Y_h shape: [1, 2, 2]
+    std::vector<float> expected_Y_h = {-0.1856241226196289f,
+                                       -0.027182353660464287f,
+                                       -0.1143757626414299f,
+                                       0.14773108065128326f};
+
+    // Y_c shape: [1, 2, 2]
+    std::vector<float> expected_Y_c = {-0.3040353059768677f,
+                                       -0.4779679775238037f,
+                                       -0.31235989928245544f,
+                                       0.3412477970123291f};
+
+    // Add inputs
+    test_case.add_input<float>(Shape{1, 1, 3, 2, 4}, X);
+    test_case.add_input<float>(Shape{1, 8, 4}, W);
+    test_case.add_input<float>(Shape{1, 8, 2}, R);
+    test_case.add_input<float>(Shape{1, 16}, B);
+
+    // Add expected outputs
+    test_case.add_expected_output<float>(Shape{3, 1, 2, 2}, expected_Y);
+    test_case.add_expected_output<float>(Shape{1, 2, 2}, expected_Y_h);
+    test_case.add_expected_output<float>(Shape{1, 2, 2}, expected_Y_c);
+
+    test_case.run_with_tolerance_as_fp(1.0e-4f);
+}
+
+// Test for LSTM with Unsqueeze preprocessing (reproduces silero_vad structure)
+// Input goes through Unsqueeze(axes=[0]) to get rank-4: [1, seq, batch, input]
+// Then LSTM applies rank reduction (Squeeze path) internally
+// W, R, B are embedded as initializers, not passed as test inputs
+OPENVINO_TEST(${BACKEND_NAME}, onnx_model_lstm_rank4_with_unsqueeze) {
+    auto model = convert_model("lstm_rank4_with_unsqueeze.onnx");
+    auto test_case = ov::test::TestCase(model, s_device);
+
+    // Input X_input shape: [3, 2, 4]
+    // Inside model: Unsqueeze(axes=[0]) -> [1, 3, 2, 4]
+    // Then LSTM reduces rank-4 back to rank-3
+    std::vector<float> X_input = {
+        -0.5138669013977051f, -1.0592135190963745f,  -0.06267909705638885f, 0.9551423192024231f,   -0.9857260584831238f,
+        0.5040464997291565f,  -0.5302576422691345f,  -0.7928728461265564f,  -0.10703036189079285f, -1.0352423191070557f,
+        -0.5536493062973022f, -1.1978778839111328f,  1.964725136756897f,    0.0352635532617569f,   -0.6997255086898804f,
+        0.21397991478443146f, -0.11232805252075195f, -0.22096960246562958f, 0.6141666769981384f,   0.7575076818466187f,
+        -0.530501127243042f,  -0.5758182406425476f,  -0.27505168318748474f, -2.3019211292266846f};
+
+    // Expected outputs from ONNX Runtime
+    std::vector<float> expected_Y = {
+        -0.041506070643663406f, 0.5078101754188538f,   -0.1956426054239273f,   -0.6245316863059998f,
+        -0.46069785952568054f,  0.16945451498031616f,  -0.036582063883543015f, -0.14519499242305756f,
+        0.03458118066191673f,   0.22937199473381042f,  -0.33998003602027893f,  -0.24845707416534424f,
+        -0.10919304192066193f,  0.041263557970523834f, -0.01278171967715025f,  -0.09499483555555344f,
+        -0.05660216882824898f,  0.4603622555732727f,   -0.6351584196090698f,   0.03479098901152611f,
+        -0.88917076587677f,     0.037890415638685226f, 0.006493804045021534f,  -0.5771874189376831f,
+        0.08443683385848999f,   -0.06232306361198425f, -0.8248671293258667f,   0.05733451619744301f,
+        -0.454610675573349f,    0.02996906079351902f,  0.0028792552184313536f, -0.045823417603969574f,
+        0.3284722566604614f,    0.4379877746105194f,   -0.22600656747817993f,  -0.107466921210289f,
+        -0.6092900633811951f,   0.4817439317703247f,   -0.06534680724143982f,  -0.039009325206279755f,
+        0.191391259431839f,     0.07052464038133621f,  -0.33953383564949036f,  0.14264167845249176f,
+        -0.7233807444572449f,   0.01202191412448883f,  0.009454132989048958f,  -0.23135091364383698f};
+
+    std::vector<float> expected_Y_h = {0.3284722566604614f,
+                                       0.4379877746105194f,
+                                       -0.22600656747817993f,
+                                       -0.107466921210289f,
+                                       -0.6092900633811951f,
+                                       0.4817439317703247f,
+                                       -0.06534680724143982f,
+                                       -0.039009325206279755f,
+                                       0.191391259431839f,
+                                       0.07052464038133621f,
+                                       -0.33953383564949036f,
+                                       0.14264167845249176f,
+                                       -0.7233807444572449f,
+                                       0.01202191412448883f,
+                                       0.009454132989048958f,
+                                       -0.23135091364383698f};
+
+    std::vector<float> expected_Y_c = {0.843663215637207f,
+                                       0.4875003397464752f,
+                                       -0.501198947429657f,
+                                       -0.2064143717288971f,
+                                       -1.8175270557403564f,
+                                       0.9995222091674805f,
+                                       -0.0911954939365387f,
+                                       -1.0586004257202148f,
+                                       0.2041035294532776f,
+                                       0.0721229761838913f,
+                                       -1.948965311050415f,
+                                       1.1799299716949463f,
+                                       -2.050872564315796f,
+                                       0.5028446316719055f,
+                                       0.1787012219429016f,
+                                       -0.24886609613895416f};
+
+    // Add input (only X, weights are embedded)
+    test_case.add_input<float>(Shape{3, 2, 4}, X_input);
+
+    // Add expected outputs
+    test_case.add_expected_output<float>(Shape{3, 1, 2, 8}, expected_Y);
+    test_case.add_expected_output<float>(Shape{1, 2, 8}, expected_Y_h);
+    test_case.add_expected_output<float>(Shape{1, 2, 8}, expected_Y_c);
+
+    test_case.run_with_tolerance_as_fp(1.0e-4f);
+}
+
 // RNNLikeSequenceOp test fixture for test setup reuse
 class GRUSequenceOp : public testing::Test {
 public: