diff --git a/Makefile b/Makefile
index 9b6552bf..5263e686 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?=0d6fb52be0c1b7e77eb855f3adc4952771c8ce4c
+CPPLLAMA_VERSION?=0a4ce786814b123096d18aadca89cd352b9e590b
 
 # gpt4all version
 GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
diff --git a/backend/cpp/llama/grpc-server.cpp b/backend/cpp/llama/grpc-server.cpp
index cb5c85f1..5de46798 100644
--- a/backend/cpp/llama/grpc-server.cpp
+++ b/backend/cpp/llama/grpc-server.cpp
@@ -458,7 +458,9 @@ struct llama_server_context
             }
         }
 
-        std::tie(model, ctx) = llama_init_from_gpt_params(params);
+        llama_init_result llama_init = llama_init_from_gpt_params(params);
+        model = llama_init.model;
+        ctx = llama_init.context;
         if (model == nullptr)
         {
             LOG_ERROR("unable to load model", {{"model", params.model}});