diff --git a/src/backend_x64/block_of_code.cpp b/src/backend_x64/block_of_code.cpp
index 2aa6210d..27a34f51 100644
--- a/src/backend_x64/block_of_code.cpp
+++ b/src/backend_x64/block_of_code.cpp
@@ -105,12 +105,6 @@ void BlockOfCode::GenRunCode() {
     align();
     run_code = getCurr<RunCodeFuncType>();
 
-    // As we currently do not emit AVX instructions, AVX-SSE transition may occur.
-    // We avoid the transition penalty by calling vzeroupper.
-    if (DoesCpuSupport(Xbyak::util::Cpu::tAVX)) {
-        vzeroupper();
-    }
-
     // This serves two purposes:
     // 1. It saves all the registers we as a callee need to save.
     // 2. It aligns the stack so that the code the JIT emits can assume
diff --git a/src/backend_x64/block_of_code.h b/src/backend_x64/block_of_code.h
index d2dd2182..d8cbe7ef 100644
--- a/src/backend_x64/block_of_code.h
+++ b/src/backend_x64/block_of_code.h
@@ -51,12 +51,6 @@ public:
         const u64 address  = reinterpret_cast<u64>(fn);
         const u64 distance = address - (getCurr<u64>() + 5);
 
-        // As we do not know if user-code is AVX or SSE, an AVX-SSE transition may occur.
-        // We avoid the transition penalty by calling vzeroupper.
-        if (DoesCpuSupport(Xbyak::util::Cpu::tAVX)) {
-            vzeroupper();
-        }
-
         if (distance >= 0x0000000080000000ULL && distance < 0xFFFFFFFF80000000ULL) {
             // Far call
             mov(rax, address);