diff --git a/CMakeLists.txt b/CMakeLists.txt
index 46c3cd5c570..90139bc1c9c 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1002,6 +1002,9 @@ if(EXECUTORCH_BUILD_EXTENSION_LLM)
     )
     set(CMAKE_POSITION_INDEPENDENT_CODE ON)
   endif()
+  set(TOKENIZERS_BUILD_HF_RUST_TOKENIZER ${EXECUTORCH_BUILD_HF_RUST_TOKENIZER})
+  set(TOKENIZERS_HF_RUST_FORMATS ${EXECUTORCH_HF_RUST_FORMATS})
+  set(TOKENIZERS_OPTIMIZE_SIZE ${EXECUTORCH_OPTIMIZE_SIZE})
   add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/extension/llm/tokenizers)
   if(EXECUTORCH_BUILD_EXTENSION_LLM_RUNNER)
     set(CMAKE_POSITION_INDEPENDENT_CODE
diff --git a/extension/llm/runner/CMakeLists.txt b/extension/llm/runner/CMakeLists.txt
index 17016b955e7..5f4579f5123 100644
--- a/extension/llm/runner/CMakeLists.txt
+++ b/extension/llm/runner/CMakeLists.txt
@@ -49,6 +49,12 @@ set(runner_deps
     executorch_core extension_module extension_tensor extension_llm_sampler
     extension_memory_allocator tokenizers::tokenizers
 )
+if(EXECUTORCH_BUILD_HF_RUST_TOKENIZER)
+  list(APPEND runner_deps tokenizers::hf_rust_tokenizer)
+  target_compile_definitions(
+    extension_llm_runner PRIVATE EXECUTORCH_USE_HF_RUST_TOKENIZER=1
+  )
+endif()
 
 # depend on arange_utils
 if(NOT TARGET kernels_util_all_deps)
diff --git a/extension/llm/runner/llm_runner_helper.cpp b/extension/llm/runner/llm_runner_helper.cpp
index 105b3f0c9f4..887039a65da 100644
--- a/extension/llm/runner/llm_runner_helper.cpp
+++ b/extension/llm/runner/llm_runner_helper.cpp
@@ -25,6 +25,9 @@
 #include <pytorch/tokenizers/sentencepiece.h>
 #include <pytorch/tokenizers/tekken.h>
 #include <pytorch/tokenizers/tiktoken.h>
+#ifdef EXECUTORCH_USE_HF_RUST_TOKENIZER
+#include <pytorch/tokenizers/rust_hf_tokenizer.h>
+#endif
 
 namespace executorch::extension::llm {
 
@@ -50,6 +53,13 @@ std::unique_ptr<tokenizers::Tokenizer> load_tokenizer(
       return tekken_tokenizer;
     }
   }
+#ifdef EXECUTORCH_USE_HF_RUST_TOKENIZER
+  auto rust_tok_tokenizer = std::make_unique<tokenizers::RustHFTokenizer>();
+  if (rust_tok_tokenizer->load(tokenizer_path) == ::tokenizers::Error::Ok) {
+    ET_LOG(Info, "Loaded Hugging Face Rust tokenizer");
+    return rust_tok_tokenizer;
+  }
+#endif
   auto json_tokenizer = std::make_unique<tokenizers::HFTokenizer>();
   if (json_tokenizer->load(tokenizer_path) == ::tokenizers::Error::Ok) {
     ET_LOG(Info, "Loaded json tokenizer");
diff --git a/extension/llm/tokenizers b/extension/llm/tokenizers
index 3fc1df71045..48970bb5640 160000
--- a/extension/llm/tokenizers
+++ b/extension/llm/tokenizers
@@ -1 +1 @@
-Subproject commit 3fc1df71045f76841ee912b31f1f176067de2e16
+Subproject commit 48970bb564014e075a0c05cda787cebcb3bbc1fd
diff --git a/tools/cmake/preset/default.cmake b/tools/cmake/preset/default.cmake
index 89e679aec7c..a63e48c8da0 100644
--- a/tools/cmake/preset/default.cmake
+++ b/tools/cmake/preset/default.cmake
@@ -100,6 +100,17 @@ define_overridable_option(
 define_overridable_option(
   EXECUTORCH_BUILD_EXTENSION_LLM "Build the LLM extension" BOOL OFF
 )
+define_overridable_option(
+  EXECUTORCH_BUILD_HF_RUST_TOKENIZER
+  "Build the opt-in Hugging Face Rust tokenizer backend" BOOL OFF
+)
+set(EXECUTORCH_HF_RUST_FORMATS
+    "json"
+    CACHE STRING "Hugging Face Rust tokenizer formats: json, tok, or json,tok"
+)
+set_property(
+  CACHE EXECUTORCH_HF_RUST_FORMATS PROPERTY STRINGS json tok "json,tok"
+)
 define_overridable_option(
   EXECUTORCH_BUILD_EXTENSION_ETDUMP_APPLE "Build the ETDump Apple extension"
   BOOL OFF