diff --git a/src/server/stream_family.cc b/src/server/stream_family.cc
index e7beb92613a9..0ea4c02bd7c0 100644
--- a/src/server/stream_family.cc
+++ b/src/server/stream_family.cc
@@ -2948,11 +2948,16 @@ void XReadGeneric2(CmdArgList args, ConnectionContext* cntx, bool read_group) {
 
   auto* tx = cntx->transaction;
 
-  // Determine if streams have entries
+  // Determine if streams have entries or any error occured
   AggregateValue<optional<facade::ErrorReply>> err;
   atomic_bool have_entries = false;
 
-  auto cb = [&](auto* tx, auto* es) {
+  // With a single shard we can call OpRead in a single hop, falling back to
+  // avoid concluding if no entries are available.
+  bool try_fastread = tx->GetUniqueShardCnt() == 1;
+  vector<RecordVec> fastread_prefetched;
+
+  auto cb = [&](auto* tx, auto* es) -> Transaction::RunnableResult {
     auto op_args = tx->GetOpArgs(es);
     for (string_view skey : tx->GetShardArgs(es->shard_id())) {
       if (auto res = HasEntries2(op_args, skey, &*opts); holds_alternative<facade::ErrorReply>(res))
@@ -2960,9 +2965,16 @@ void XReadGeneric2(CmdArgList args, ConnectionContext* cntx, bool read_group) {
       else if (holds_alternative<bool>(res) && get<bool>(res))
         have_entries.store(true, memory_order_relaxed);
     }
+
+    if (try_fastread) {
+      if (have_entries.load(memory_order_relaxed))
+        fastread_prefetched = OpRead(tx->GetOpArgs(es), tx->GetShardArgs(es->shard_id()), *opts);
+      else
+        return {OpStatus::OK, Transaction::RunnableResult::AVOID_CONCLUDING};
+    }
     return OpStatus::OK;
   };
-  tx->Execute(cb, false);
+  tx->Execute(cb, try_fastread);
 
   if (err) {
     tx->Conclude();
@@ -2973,13 +2985,17 @@ void XReadGeneric2(CmdArgList args, ConnectionContext* cntx, bool read_group) {
     return XReadBlock(&*opts, cntx);
 
   vector<vector<RecordVec>> xread_resp;
-  xread_resp.resize(shard_set->size());
-  auto read_cb = [&](Transaction* t, EngineShard* shard) {
-    ShardId sid = shard->shard_id();
-    xread_resp[sid] = OpRead(t->GetOpArgs(shard), t->GetShardArgs(sid), *opts);
-    return OpStatus::OK;
-  };
-  tx->Execute(std::move(read_cb), true);
+  if (try_fastread && have_entries.load(memory_order_relaxed)) {
+    xread_resp = {std::move(fastread_prefetched)};
+  } else {
+    xread_resp.resize(shard_set->size());
+    auto read_cb = [&](Transaction* t, EngineShard* shard) {
+      ShardId sid = shard->shard_id();
+      xread_resp[sid] = OpRead(t->GetOpArgs(shard), t->GetShardArgs(sid), *opts);
+      return OpStatus::OK;
+    };
+    tx->Execute(std::move(read_cb), true);
+  }
 
   // Count number of streams and merge final results in correct order
   int resolved_streams = 0;
diff --git a/src/server/stream_family_test.cc b/src/server/stream_family_test.cc
index fd1cadfe8f02..bedde9b80df0 100644
--- a/src/server/stream_family_test.cc
+++ b/src/server/stream_family_test.cc
@@ -133,7 +133,7 @@ TEST_F(StreamFamilyTest, XRead) {
   // Receive all records from a single stream, in a single hop
   auto resp = Run({"xread", "streams", "foo", "0"});
   EXPECT_THAT(resp.GetVec(), ElementsAre("foo", ArrLen(3)));
-  // EXPECT_EQ(GetMetrics().shard_stats.tx_optimistic_total, 5u); todo temporary disabled
+  EXPECT_EQ(GetMetrics().shard_stats.tx_optimistic_total, 5u);
 
   // Receive all records from both streams.
   resp = Run({"xread", "streams", "foo", "bar", "0", "0"});