diff --git a/Makefile b/Makefile
index 297938ae4c46..f7b4abb561d1 100644
--- a/Makefile
+++ b/Makefile
@@ -8,7 +8,7 @@ DETECT_LIBS?=true
 # llama.cpp versions
 GOLLAMA_REPO?=https://github.com/go-skynet/go-llama.cpp
 GOLLAMA_VERSION?=2b57a8ae43e4699d3dc5d1496a1ccd42922993be
-CPPLLAMA_VERSION?=081fe431aa8fb6307145c4feb3eed4f48cab19f8
+CPPLLAMA_VERSION?=b841d0740855c5af1344a81f261139a45a2b39ee
 
 # gpt4all version
 GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
diff --git a/backend/cpp/llama/CMakeLists.txt b/backend/cpp/llama/CMakeLists.txt
index 031e49643fbb..1c68e60b42e3 100644
--- a/backend/cpp/llama/CMakeLists.txt
+++ b/backend/cpp/llama/CMakeLists.txt
@@ -75,11 +75,24 @@ add_library(hw_grpc_proto
   ${hw_proto_hdrs} )
 
 add_executable(${TARGET} grpc-server.cpp utils.hpp json.hpp)
-target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
-  absl::flags_parse
-  gRPC::${_REFLECTION}
-  gRPC::${_GRPC_GRPCPP}
-  protobuf::${_PROTOBUF_LIBPROTOBUF})
+
+# Conditionally link SYCL to grpc-server
+# https://github.com/ggerganov/llama.cpp/issues/8665
+if ( DEFINED ENV{ONEAPI_ROOT})
+    target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
+      absl::flags_parse
+      gRPC::${_REFLECTION}
+      gRPC::${_GRPC_GRPCPP}
+      protobuf::${_PROTOBUF_LIBPROTOBUF}
+      sycl)
+else()
+    target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
+      absl::flags_parse
+      gRPC::${_REFLECTION}
+      gRPC::${_GRPC_GRPCPP}
+      protobuf::${_PROTOBUF_LIBPROTOBUF})
+endif()
+
 target_compile_features(${TARGET} PRIVATE cxx_std_11)
 if(TARGET BUILD_INFO)
   add_dependencies(${TARGET} BUILD_INFO)
diff --git a/backend/cpp/llama/CMakeLists.txt.rpc-8662 b/backend/cpp/llama/CMakeLists.txt.rpc-8662
new file mode 100644
index 000000000000..82bbedcaa838
--- /dev/null
+++ b/backend/cpp/llama/CMakeLists.txt.rpc-8662
@@ -0,0 +1,8 @@
+# https://github.com/ggerganov/llama.cpp/issues/8665
+
+add_executable(rpc-server rpc-server.cpp)
+if ( DEFINED ENV{ONEAPI_ROOT})
+target_link_libraries(rpc-server PRIVATE ggml llama sycl)
+else()
+target_link_libraries(rpc-server PRIVATE ggml llama)
+endif()
\ No newline at end of file
diff --git a/backend/cpp/llama/prepare.sh b/backend/cpp/llama/prepare.sh
index 6c00f27caa38..c1c94ce6c9d6 100644
--- a/backend/cpp/llama/prepare.sh
+++ b/backend/cpp/llama/prepare.sh
@@ -17,4 +17,7 @@ cp -rfv llama.cpp/examples/llava/clip.h llama.cpp/examples/grpc-server/clip.h
 cp -rfv llama.cpp/examples/llava/llava.cpp llama.cpp/examples/grpc-server/llava.cpp
 echo '#include "llama.h"' > llama.cpp/examples/grpc-server/llava.h
 cat llama.cpp/examples/llava/llava.h >> llama.cpp/examples/grpc-server/llava.h
-cp -rfv llama.cpp/examples/llava/clip.cpp llama.cpp/examples/grpc-server/clip.cpp
\ No newline at end of file
+cp -rfv llama.cpp/examples/llava/clip.cpp llama.cpp/examples/grpc-server/clip.cpp
+
+# https://github.com/ggerganov/llama.cpp/issues/8665
+cp -rfv CMakeLists.txt.rpc-8662 llama.cpp/examples/rpc/CMakeLists.txt
\ No newline at end of file