diff --git a/common/arg.cpp b/common/arg.cpp
index ec0a2f015e..f2675f842a 100644
--- a/common/arg.cpp
+++ b/common/arg.cpp
@@ -1295,7 +1295,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
         [](common_params & params) {
             params.kv_unified = true;
         }
-    ).set_env("LLAMA_ARG_KV_UNIFIED").set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_PERPLEXITY}));
+    ).set_env("LLAMA_ARG_KV_UNIFIED").set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_PERPLEXITY, LLAMA_EXAMPLE_BATCHED}));
     add_opt(common_arg(
         {"--context-shift"},
         {"--no-context-shift"},
diff --git a/common/common.h b/common/common.h
index 7794c0268b..b3ac04c4ae 100644
--- a/common/common.h
+++ b/common/common.h
@@ -80,6 +80,7 @@ int32_t cpu_get_num_math();
 //
 
 enum llama_example {
+    LLAMA_EXAMPLE_BATCHED,
     LLAMA_EXAMPLE_DEBUG,
     LLAMA_EXAMPLE_COMMON,
     LLAMA_EXAMPLE_SPECULATIVE,
diff --git a/examples/batched/batched.cpp b/examples/batched/batched.cpp
index 6b134b4f6f..6875038770 100644
--- a/examples/batched/batched.cpp
+++ b/examples/batched/batched.cpp
@@ -21,7 +21,7 @@ int main(int argc, char ** argv) {
     params.prompt = "Hello my name is";
     params.n_predict = 32;
 
-    if (!common_params_parse(argc, argv, params, LLAMA_EXAMPLE_COMMON, print_usage)) {
+    if (!common_params_parse(argc, argv, params, LLAMA_EXAMPLE_BATCHED, print_usage)) {
         return 1;
     }