diff --git a/gen/gen_code.cpp b/gen/gen_code.cpp
index 52674c2..24bd3aa 100644
--- a/gen/gen_code.cpp
+++ b/gen/gen_code.cpp
@@ -1520,14 +1520,12 @@
 	}
 	// mov
 	{
-		printf("void vmovd(const Xmm& x, const Reg32& reg) { opAVX_X_X_XM(x, xm0, Xmm(reg.getIdx()), T_0F | T_66 | T_W0, 0x6E); }\n");
-		printf("void vmovd(const Xmm& x, const Address& addr) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_66 | T_W0, 0x6E); }\n");
-		printf("void vmovd(const Reg32& reg, const Xmm& x) { opAVX_X_X_XM(x, xm0, Xmm(reg.getIdx()), T_0F | T_66 | T_W0, 0x7E); }\n");
-		printf("void vmovd(const Address& addr, const Xmm& x) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_66 | T_W0, 0x7E); }\n");
+		printf("void vmovd(const Xmm& x, const Operand& op) { if (!op.isREG(32) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x, xm0, op, T_0F | T_66 | T_W0 | T_EVEX, 0x6E); }\n");
+		printf("void vmovd(const Operand& op, const Xmm& x) { if (!op.isREG(32) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x, xm0, op, T_0F | T_66 | T_W0 | T_EVEX, 0x7E); }\n");
 
-		printf("void vmovq(const Xmm& x, const Address& addr) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_F3, 0x7E); }\n");
-		printf("void vmovq(const Address& addr, const Xmm& x) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_66, 0xD6); }\n");
-		printf("void vmovq(const Xmm& x1, const Xmm& x2) { opAVX_X_X_XM(x1, xm0, x2, T_0F | T_F3, 0x7E); }\n");
+		printf("void vmovq(const Xmm& x, const Address& addr) { int type, code; if (x.getIdx() < 16) { type = T_0F | T_F3; code = 0x7E; } else { type = T_0F | T_66 | T_EVEX | T_EW1; code = 0x6E; } opAVX_X_X_XM(x, xm0, addr, type, code); }\n");
+		printf("void vmovq(const Address& addr, const Xmm& x) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_66 | T_EVEX | T_EW1, x.getIdx() < 16 ? 0xD6 : 0x7E); }\n");
+		printf("void vmovq(const Xmm& x1, const Xmm& x2) { opAVX_X_X_XM(x1, xm0, x2, T_0F | T_F3 | T_EVEX | T_EW1, 0x7E); }\n");
 
 		printf("void vmovhlps(const Xmm& x1, const Xmm& x2, const Operand& op = Operand()) { if (!op.isNone() && !op.isXMM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x1, x2, op, T_0F, 0x12); }\n");
 		printf("void vmovlhps(const Xmm& x1, const Xmm& x2, const Operand& op = Operand()) { if (!op.isNone() && !op.isXMM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x1, x2, op, T_0F, 0x16); }\n");
@@ -1573,9 +1571,8 @@
 	{
 
 		printf("#ifdef XBYAK64\n");
-		printf("void vmovq(const Xmm& x, const Reg64& reg) { opAVX_X_X_XM(x, xm0, Xmm(reg.getIdx()), T_0F | T_66 | T_W1, 0x6E); }\n");
-		printf("void vmovq(const Reg64& reg, const Xmm& x) { opAVX_X_X_XM(x, xm0, Xmm(reg.getIdx()), T_0F | T_66 | T_W1, 0x7E); }\n");
-
+		printf("void vmovq(const Xmm& x, const Reg64& r) { opAVX_X_X_XM(x, xm0, Xmm(r.getIdx()), T_66 | T_0F | T_W1 | T_EVEX | T_EW1, 0x6E); }\n");
+		printf("void vmovq(const Reg64& r, const Xmm& x) { opAVX_X_X_XM(x, xm0, Xmm(r.getIdx()), T_66 | T_0F | T_W1 | T_EVEX | T_EW1, 0x7E); }\n");
 		printf("void vpextrq(const Operand& op, const Xmm& x, uint8 imm) { if (!op.isREG(64) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XMcvt(x, xm0, op, !op.isMEM(), Operand::XMM, T_0F3A | T_66 | T_W1, 0x16, imm); }\n");
 
 		printf("void vpinsrq(const Xmm& x1, const Xmm& x2, const Operand& op, uint8 imm) { if (!op.isREG(64) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XMcvt(x1, x2, op, !op.isMEM(), Operand::XMM, T_0F3A | T_66 | T_W1, 0x22, imm); }\n");
diff --git a/test/make_nm.cpp b/test/make_nm.cpp
index 2446a84..2c74346 100644
--- a/test/make_nm.cpp
+++ b/test/make_nm.cpp
@@ -2590,6 +2590,15 @@
 			put(name, ZMM, MEM);
 		}
 	}
+	void put_vmov()
+	{
+#ifdef XBYAK64
+		put("vmovd", _XMM3, MEM|REG32);
+		put("vmovd", MEM|REG32, _XMM3);
+		put("vmovq", _XMM3, MEM|REG64|XMM);
+		put("vmovq", MEM|REG64|XMM, _XMM3);
+#endif
+	}
 	void putAVX512()
 	{
 		putOpmask();
@@ -2597,6 +2606,7 @@
 		putCmpK();
 		putBroadcast();
 		putAVX512_M_X();
+		put_vmov();
 	}
 #endif
 };
diff --git a/xbyak/xbyak_mnemonic.h b/xbyak/xbyak_mnemonic.h
index 592e0e2..5845639 100644
--- a/xbyak/xbyak_mnemonic.h
+++ b/xbyak/xbyak_mnemonic.h
@@ -1454,13 +1454,11 @@
 void vblendvps(const Xmm& x1, const Operand& op, const Xmm& x4) { vblendvps(x1, x1, op, x4); }
 void vpblendvb(const Xmm& x1, const Xmm& x2, const Operand& op, const Xmm& x4) { opAVX_X_X_XM(x1, x2, op, T_0F3A | T_66 | T_YMM, 0x4C, x4.getIdx() << 4); }
 void vpblendvb(const Xmm& x1, const Operand& op, const Xmm& x4) { vpblendvb(x1, x1, op, x4); }
-void vmovd(const Xmm& x, const Reg32& reg) { opAVX_X_X_XM(x, xm0, Xmm(reg.getIdx()), T_0F | T_66 | T_W0, 0x6E); }
-void vmovd(const Xmm& x, const Address& addr) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_66 | T_W0, 0x6E); }
-void vmovd(const Reg32& reg, const Xmm& x) { opAVX_X_X_XM(x, xm0, Xmm(reg.getIdx()), T_0F | T_66 | T_W0, 0x7E); }
-void vmovd(const Address& addr, const Xmm& x) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_66 | T_W0, 0x7E); }
-void vmovq(const Xmm& x, const Address& addr) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_F3, 0x7E); }
-void vmovq(const Address& addr, const Xmm& x) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_66, 0xD6); }
-void vmovq(const Xmm& x1, const Xmm& x2) { opAVX_X_X_XM(x1, xm0, x2, T_0F | T_F3, 0x7E); }
+void vmovd(const Xmm& x, const Operand& op) { if (!op.isREG(32) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x, xm0, op, T_0F | T_66 | T_W0 | T_EVEX, 0x6E); }
+void vmovd(const Operand& op, const Xmm& x) { if (!op.isREG(32) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x, xm0, op, T_0F | T_66 | T_W0 | T_EVEX, 0x7E); }
+void vmovq(const Xmm& x, const Address& addr) { int type, code; if (x.getIdx() < 16) { type = T_0F | T_F3; code = 0x7E; } else { type = T_0F | T_66 | T_EVEX | T_EW1; code = 0x6E; } opAVX_X_X_XM(x, xm0, addr, type, code); }
+void vmovq(const Address& addr, const Xmm& x) { opAVX_X_X_XM(x, xm0, addr, T_0F | T_66 | T_EVEX | T_EW1, x.getIdx() < 16 ? 0xD6 : 0x7E); }
+void vmovq(const Xmm& x1, const Xmm& x2) { opAVX_X_X_XM(x1, xm0, x2, T_0F | T_F3 | T_EVEX | T_EW1, 0x7E); }
 void vmovhlps(const Xmm& x1, const Xmm& x2, const Operand& op = Operand()) { if (!op.isNone() && !op.isXMM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x1, x2, op, T_0F, 0x12); }
 void vmovlhps(const Xmm& x1, const Xmm& x2, const Operand& op = Operand()) { if (!op.isNone() && !op.isXMM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x1, x2, op, T_0F, 0x16); }
 void vmovmskpd(const Reg& r, const Xmm& x) { if (!r.isBit(i32e)) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XM(x.isXMM() ? Xmm(r.getIdx()) : Ymm(r.getIdx()), x.isXMM() ? xm0 : ym0, x, T_0F | T_66 | T_W0 | T_YMM, 0x50); }
@@ -1489,8 +1487,8 @@
 void vcvtph2ps(const Xmm& x, const Operand& op) { if (!op.isMEM() && !op.isXMM()) throw Error(ERR_BAD_COMBINATION); opVex(x, 0, op, T_0F38 | T_66 | T_W0, 0x13); }
 void vcvtps2ph(const Operand& op, const Xmm& x, uint8 imm) { if (!op.isMEM() && !op.isXMM()) throw Error(ERR_BAD_COMBINATION); opVex(x, 0, op, T_0F3A | T_66 | T_W0, 0x1d, imm); }
 #ifdef XBYAK64
-void vmovq(const Xmm& x, const Reg64& reg) { opAVX_X_X_XM(x, xm0, Xmm(reg.getIdx()), T_0F | T_66 | T_W1, 0x6E); }
-void vmovq(const Reg64& reg, const Xmm& x) { opAVX_X_X_XM(x, xm0, Xmm(reg.getIdx()), T_0F | T_66 | T_W1, 0x7E); }
+void vmovq(const Xmm& x, const Reg64& r) { opAVX_X_X_XM(x, xm0, Xmm(r.getIdx()), T_66 | T_0F | T_W1 | T_EVEX | T_EW1, 0x6E); }
+void vmovq(const Reg64& r, const Xmm& x) { opAVX_X_X_XM(x, xm0, Xmm(r.getIdx()), T_66 | T_0F | T_W1 | T_EVEX | T_EW1, 0x7E); }
 void vpextrq(const Operand& op, const Xmm& x, uint8 imm) { if (!op.isREG(64) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XMcvt(x, xm0, op, !op.isMEM(), Operand::XMM, T_0F3A | T_66 | T_W1, 0x16, imm); }
 void vpinsrq(const Xmm& x1, const Xmm& x2, const Operand& op, uint8 imm) { if (!op.isREG(64) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XMcvt(x1, x2, op, !op.isMEM(), Operand::XMM, T_0F3A | T_66 | T_W1, 0x22, imm); }
 void vpinsrq(const Xmm& x, const Operand& op, uint8 imm) { if (!op.isREG(64) && !op.isMEM()) throw Error(ERR_BAD_COMBINATION); opAVX_X_X_XMcvt(x, x, op, !op.isMEM(), Operand::XMM, T_0F3A | T_66 | T_W1, 0x22, imm); }