diff --git a/src/frontend/A32/decoder/thumb32.inc b/src/frontend/A32/decoder/thumb32.inc
index 3b58d50e..8b8b8a99 100644
--- a/src/frontend/A32/decoder/thumb32.inc
+++ b/src/frontend/A32/decoder/thumb32.inc
@@ -275,7 +275,7 @@ INST(thumb32_MLS,            "MLS",                      "111110110000nnnnaaaadd
 //INST(thumb32_SMMUL,          "SMMUL",                    "111110110101----1111----000-----")
 //INST(thumb32_SMMLA,          "SMMLA",                    "111110110101------------000-----")
 //INST(thumb32_SMMLS,          "SMMLS",                    "111110110110------------000-----")
-//INST(thumb32_USAD8,          "USAD8",                    "111110110111----1111----0000----")
+INST(thumb32_USAD8,          "USAD8",                    "111110110111nnnn1111dddd0000mmmm")
 //INST(thumb32_USADA8,         "USADA8",                   "111110110111------------0000----")
 
 // Long Multiply, Long Multiply Accumulate, and Divide
diff --git a/src/frontend/A32/translate/impl/thumb32_multiply.cpp b/src/frontend/A32/translate/impl/thumb32_multiply.cpp
index 51daa8fd..2a25be52 100644
--- a/src/frontend/A32/translate/impl/thumb32_multiply.cpp
+++ b/src/frontend/A32/translate/impl/thumb32_multiply.cpp
@@ -48,4 +48,17 @@ bool ThumbTranslatorVisitor::thumb32_MUL(Reg n, Reg d, Reg m) {
     return true;
 }
 
+bool ThumbTranslatorVisitor::thumb32_USAD8(Reg n, Reg d, Reg m) {
+    if (d == Reg::PC || n == Reg::PC || m == Reg::PC) {
+        return UnpredictableInstruction();
+    }
+
+    const auto reg_m = ir.GetRegister(m);
+    const auto reg_n = ir.GetRegister(n);
+    const auto result = ir.PackedAbsDiffSumS8(reg_n, reg_m);
+
+    ir.SetRegister(d, result);
+    return true;
+}
+
 } // namespace Dynarmic::A32
diff --git a/src/frontend/A32/translate/impl/translate_thumb.h b/src/frontend/A32/translate/impl/translate_thumb.h
index e1cd29b6..9fbb3092 100644
--- a/src/frontend/A32/translate/impl/translate_thumb.h
+++ b/src/frontend/A32/translate/impl/translate_thumb.h
@@ -132,6 +132,7 @@ struct ThumbTranslatorVisitor final {
     bool thumb32_MLA(Reg n, Reg a, Reg d, Reg m);
     bool thumb32_MLS(Reg n, Reg a, Reg d, Reg m);
     bool thumb32_MUL(Reg n, Reg d, Reg m);
+    bool thumb32_USAD8(Reg n, Reg d, Reg m);
 
     // thumb32 parallel add/sub instructions
     bool thumb32_SADD8(Reg n, Reg d, Reg m);