diff --git a/vta/hardware/xilinx/src/vta.cc b/vta/hardware/xilinx/src/vta.cc
index da8ea7435a90..cb6b7b76cd11 100644
--- a/vta/hardware/xilinx/src/vta.cc
+++ b/vta/hardware/xilinx/src/vta.cc
@@ -586,19 +586,29 @@ void vta(
 
   // Instantiate temporary instruction queues (used for peeking)
   hls::stream<insn_T> tmp_load_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=tmp_load_queue)
   hls::stream<insn_T> tmp_gemm_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=tmp_gemm_queue)
   hls::stream<insn_T> tmp_store_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=tmp_store_queue)
 
   // Instatiate physical instruction queues
   hls::stream<insn_T> load_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=load_queue)
   hls::stream<insn_T> gemm_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=gemm_queue)
   hls::stream<insn_T> store_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=store_queue)
 
   // Dependence queues
   hls::stream<bool> l2g_dep_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=l2g_dep_queue)
   hls::stream<bool> s2g_dep_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=s2g_dep_queue)
   hls::stream<bool> g2l_dep_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=g2s_dep_queue)
   hls::stream<bool> g2s_dep_queue;
+  PRAGMA_HLS(HLS stream depth=STREAM_IN_DEPTH variable=g2s_dep_queue)
 
   // Instantiate memories
   inp_vec_T inp_mem[VTA_INP_BUFF_DEPTH][VTA_BATCH];
diff --git a/vta/hardware/xilinx/src/vta.h b/vta/hardware/xilinx/src/vta.h
index 6480debbe4f9..1395d5eaba8e 100644
--- a/vta/hardware/xilinx/src/vta.h
+++ b/vta/hardware/xilinx/src/vta.h
@@ -113,6 +113,13 @@ typedef ap_int<VTA_ALUOP_IMM_BIT_WIDTH> aluop_imm_T;
 /* \typedef aluop_opcode_T ALU operation shift immediate datatype*/
 typedef ap_int<VTA_LOG_ACC_WIDTH> aluop_sh_imm_T;
 
+/*!
+* Define HLS stream depth 
+*/
+#define PRAGMA_SUB(x) _Pragma (#x)
+#define PRAGMA_HLS(x) PRAGMA_SUB(x)
+#define STREAM_IN_DEPTH 8
+
 /*!
 * \brief Fetch module.
 *   Reads in \a insn_count instructions via DMA and pushes them to the