diff --git a/src/lib/arch/aarch64/Instruction_execute.cc b/src/lib/arch/aarch64/Instruction_execute.cc
index 28a254679..558ebc052 100644
--- a/src/lib/arch/aarch64/Instruction_execute.cc
+++ b/src/lib/arch/aarch64/Instruction_execute.cc
@@ -7076,6 +7076,29 @@ void Instruction::execute() {
         results_[0] = vecZip<uint8_t, 8>(sourceValues_, true);
         break;
       }
+      case Opcode::AArch64_ZIP_VG4_4Z4Z_S: {  // zip {zd1.s - zd4.s}, {zn1.s -
+                                              // zn4.s}
+        const uint32_t* zn[4];
+        zn[0] = sourceValues_[0].getAsVector<uint32_t>();
+        zn[1] = sourceValues_[1].getAsVector<uint32_t>();
+        zn[2] = sourceValues_[2].getAsVector<uint32_t>();
+        zn[3] = sourceValues_[3].getAsVector<uint32_t>();
+
+        const uint16_t quads = VL_bits / (32 * 4);
+
+        uint32_t out[4][64] = {{0}, {0}, {0}, {0}};
+        for (int r = 0; r < 4; r++) {
+          const uint16_t base = r * quads;
+          for (int q = 0; q < quads; q++) {
+            out[r][4 * q] = zn[0][base + q];
+            out[r][4 * q + 1] = zn[1][base + q];
+            out[r][4 * q + 2] = zn[2][base + q];
+            out[r][4 * q + 3] = zn[3][base + q];
+          }
+          results_[r] = RegisterValue(out[r], 256);
+        }
+        break;
+      }
       case Opcode::AArch64_ZERO_M: {  // zero {mask}
         // SME
         // Not in right context mode. Raise exception
diff --git a/test/regression/aarch64/instructions/sve.cc b/test/regression/aarch64/instructions/sve.cc
index 16a966d00..f9699593f 100644
--- a/test/regression/aarch64/instructions/sve.cc
+++ b/test/regression/aarch64/instructions/sve.cc
@@ -9145,13 +9145,26 @@ TEST_P(InstSve, zip) {
     zip1 z16.b, z12.b, z13.b
     zip2 z17.b, z14.b, z15.b
   )");
-
   CHECK_NEON(4, double, fillNeon<double>({0.5, -0.5}, VL / 8));
   CHECK_NEON(5, double, fillNeon<double>({0.75, -0.75}, VL / 8));
   CHECK_NEON(10, float, fillNeon<float>({0.5, -0.75}, VL / 8));
   CHECK_NEON(11, float, fillNeon<float>({-0.5, 0.75}, VL / 8));
   CHECK_NEON(16, int8_t, fillNeon<int8_t>({1, -2}, VL / 8));
   CHECK_NEON(17, int8_t, fillNeon<int8_t>({-1, 2}, VL / 8));
+
+  // Multi-vector
+  RUN_AARCH64(R"(
+    #32-bit
+    dup z0.s, #5
+    dup z1.s, #6
+    dup z2.s, #7
+    dup z3.s, #8
+    zip {z4.s - z7.s}, {z0.s - z3.s}
+  )");
+  CHECK_NEON(4, uint32_t, fillNeon<uint32_t>({5, 6, 7, 8}, VL / 8));
+  CHECK_NEON(5, uint32_t, fillNeon<uint32_t>({5, 6, 7, 8}, VL / 8));
+  CHECK_NEON(6, uint32_t, fillNeon<uint32_t>({5, 6, 7, 8}, VL / 8));
+  CHECK_NEON(7, uint32_t, fillNeon<uint32_t>({5, 6, 7, 8}, VL / 8));
 }
 
 TEST_P(InstSve, psel) {