diff --git a/src/backend_x64/emit_x64_vector_floating_point.cpp b/src/backend_x64/emit_x64_vector_floating_point.cpp
index de24a168..a16f2d91 100644
--- a/src/backend_x64/emit_x64_vector_floating_point.cpp
+++ b/src/backend_x64/emit_x64_vector_floating_point.cpp
@@ -614,6 +614,23 @@ void EmitX64::EmitFPVectorPairedAddLower64(EmitContext& ctx, IR::Inst* inst) {
     });
 }
 
+template<typename FPT>
+static void EmitRecipEstimate(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst) {
+    EmitTwoOpFallback(code, ctx, inst, [](VectorArray<FPT>& result, const VectorArray<FPT>& operand, FP::FPCR fpcr, FP::FPSR& fpsr) {
+        for (size_t i = 0; i < result.size(); i++) {
+            result[i] = FP::FPRecipEstimate<FPT>(operand[i], fpcr, fpsr);
+        }
+    });
+}
+
+void EmitX64::EmitFPVectorRecipEstimate32(EmitContext& ctx, IR::Inst* inst) {
+    EmitRecipEstimate<u32>(code, ctx, inst);
+}
+
+void EmitX64::EmitFPVectorRecipEstimate64(EmitContext& ctx, IR::Inst* inst) {
+    EmitRecipEstimate<u64>(code, ctx, inst);
+}
+
 template<typename FPT>
 static void EmitRSqrtEstimate(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst) {
     EmitTwoOpFallback(code, ctx, inst, [](VectorArray<FPT>& result, const VectorArray<FPT>& operand, FP::FPCR fpcr, FP::FPSR& fpsr) {
diff --git a/src/frontend/ir/ir_emitter.cpp b/src/frontend/ir/ir_emitter.cpp
index fc69608e..9a060056 100644
--- a/src/frontend/ir/ir_emitter.cpp
+++ b/src/frontend/ir/ir_emitter.cpp
@@ -1749,6 +1749,17 @@ U128 IREmitter::FPVectorPairedAddLower(size_t esize, const U128& a, const U128&
     return {};
 }
 
+U128 IREmitter::FPVectorRecipEstimate(size_t esize, const U128& a) {
+    switch (esize) {
+    case 32:
+        return Inst<U128>(Opcode::FPVectorRecipEstimate32, a);
+    case 64:
+        return Inst<U128>(Opcode::FPVectorRecipEstimate64, a);
+    }
+    UNREACHABLE();
+    return {};
+}
+
 U128 IREmitter::FPVectorRSqrtEstimate(size_t esize, const U128& a) {
     switch (esize) {
     case 32:
diff --git a/src/frontend/ir/ir_emitter.h b/src/frontend/ir/ir_emitter.h
index 85747d1b..da013cd2 100644
--- a/src/frontend/ir/ir_emitter.h
+++ b/src/frontend/ir/ir_emitter.h
@@ -305,6 +305,7 @@ public:
     U128 FPVectorNeg(size_t esize, const U128& a);
     U128 FPVectorPairedAdd(size_t esize, const U128& a, const U128& b);
     U128 FPVectorPairedAddLower(size_t esize, const U128& a, const U128& b);
+    U128 FPVectorRecipEstimate(size_t esize, const U128& a);
     U128 FPVectorRSqrtEstimate(size_t esize, const U128& a);
     U128 FPVectorRSqrtStepFused(size_t esize, const U128& a, const U128& b);
     U128 FPVectorSub(size_t esize, const U128& a, const U128& b);
diff --git a/src/frontend/ir/opcodes.inc b/src/frontend/ir/opcodes.inc
index 1f115310..7ec26325 100644
--- a/src/frontend/ir/opcodes.inc
+++ b/src/frontend/ir/opcodes.inc
@@ -452,6 +452,8 @@ OPCODE(FPVectorPairedAddLower32,                T::U128,        T::U128,
 OPCODE(FPVectorPairedAddLower64,                T::U128,        T::U128,        T::U128                         )
 OPCODE(FPVectorPairedAdd32,                     T::U128,        T::U128,        T::U128                         )
 OPCODE(FPVectorPairedAdd64,                     T::U128,        T::U128,        T::U128                         )
+OPCODE(FPVectorRecipEstimate32,                 T::U128,        T::U128                                         )
+OPCODE(FPVectorRecipEstimate64,                 T::U128,        T::U128                                         )
 OPCODE(FPVectorRSqrtEstimate32,                 T::U128,        T::U128                                         )
 OPCODE(FPVectorRSqrtEstimate64,                 T::U128,        T::U128                                         )
 OPCODE(FPVectorRSqrtStepFused32,                T::U128,        T::U128,        T::U128                         )