diff --git a/examples/common.cpp b/examples/common.cpp
index 35be2b5aa..4ee1ea79a 100644
--- a/examples/common.cpp
+++ b/examples/common.cpp
@@ -556,7 +556,7 @@ std::vector<llama_token> llama_tokenize(struct llama_context * ctx, const std::s
     return res;
 }
 
-std::tuple<struct llama_model *, struct llama_context *, struct llama_context_params> llama_init_from_gpt_params(const gpt_params & params) {
+struct llama_context_params llama_get_context_params_from_gpt_params(const gpt_params & params) {
     auto lparams = llama_context_default_params();
 
     lparams.n_ctx        = params.n_ctx;
@@ -572,17 +572,23 @@ std::tuple<struct llama_model *, struct llama_context *, struct llama_context_pa
     lparams.logits_all   = params.perplexity;
     lparams.embedding    = params.embedding;
 
+    return lparams;
+}
+
+std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_params(const gpt_params & params) {
+    auto lparams = llama_get_context_params_from_gpt_params(params);
+
     llama_model * model  = llama_load_model_from_file(params.model.c_str(), lparams);
     if (model == NULL) {
         fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, params.model.c_str());
-        return std::make_tuple(nullptr, nullptr, lparams);
+        return std::make_tuple(nullptr, nullptr);
     }
 
     llama_context * lctx = llama_new_context_with_model(model, lparams);
     if (lctx == NULL) {
         fprintf(stderr, "%s: error: failed to create context with model '%s'\n", __func__, params.model.c_str());
         llama_free_model(model);
-        return std::make_tuple(nullptr, nullptr, lparams);
+        return std::make_tuple(nullptr, nullptr);
     }
 
     if (!params.lora_adapter.empty()) {
@@ -594,11 +600,11 @@ std::tuple<struct llama_model *, struct llama_context *, struct llama_context_pa
             fprintf(stderr, "%s: error: failed to apply lora adapter\n", __func__);
             llama_free(lctx);
             llama_free_model(model);
-            return std::make_tuple(nullptr, nullptr, lparams);
+            return std::make_tuple(nullptr, nullptr);
         }
     }
 
-    return std::make_tuple(model, lctx, lparams);
+    return std::make_tuple(model, lctx);
 }
 
 void console_init(console_state & con_st) {
diff --git a/examples/common.h b/examples/common.h
index bed576438..00cb6888a 100644
--- a/examples/common.h
+++ b/examples/common.h
@@ -104,7 +104,8 @@ std::vector<llama_token> llama_tokenize(struct llama_context * ctx, const std::s
 // Model utils
 //
 
-std::tuple<struct llama_model *, struct llama_context *, struct llama_context_params> llama_init_from_gpt_params(const gpt_params & params);
+std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_params(const gpt_params & params);
+struct llama_context_params llama_get_context_params_from_gpt_params(const gpt_params & params);
 
 //
 // Console utils
diff --git a/examples/embd-input/embd-input-lib.cpp b/examples/embd-input/embd-input-lib.cpp
index 576ac0af0..5fa4942be 100644
--- a/examples/embd-input/embd-input-lib.cpp
+++ b/examples/embd-input/embd-input-lib.cpp
@@ -42,7 +42,7 @@ struct MyModel* create_mymodel(int argc, char ** argv) {
     g_ctx = &ctx;
 
     // load the model and apply lora adapter, if any
-    std::tie(model, ctx, std::ignore) = llama_init_from_gpt_params(params);
+    std::tie(model, ctx) = llama_init_from_gpt_params(params);
     if (model == NULL) {
         fprintf(stderr, "%s: error: unable to load model\n", __func__);
         return nullptr;
diff --git a/examples/embedding/embedding.cpp b/examples/embedding/embedding.cpp
index 7b1135e6a..03e801c2a 100644
--- a/examples/embedding/embedding.cpp
+++ b/examples/embedding/embedding.cpp
@@ -41,7 +41,7 @@ int main(int argc, char ** argv) {
     llama_context * ctx;
 
     // load the model
-    std::tie(model, ctx, std::ignore) = llama_init_from_gpt_params(params);
+    std::tie(model, ctx) = llama_init_from_gpt_params(params);
     if (model == NULL) {
         fprintf(stderr, "%s: error: unable to load model\n", __func__);
         return 1;
diff --git a/examples/main/main.cpp b/examples/main/main.cpp
index 88262c920..dbc384f52 100644
--- a/examples/main/main.cpp
+++ b/examples/main/main.cpp
@@ -124,12 +124,12 @@ int main(int argc, char ** argv) {
     llama_model * model;
     llama_context * ctx;
     llama_context * guidance_ctx = NULL;
-    struct llama_context_params lparams;
     g_ctx = &ctx;
 
     // load the model and apply lora adapter, if any
-    std::tie(model, ctx, lparams) = llama_init_from_gpt_params(params);
+    std::tie(model, ctx) = llama_init_from_gpt_params(params);
     if (params.cfg_scale > 1.f) {
+        struct llama_context_params lparams = llama_get_context_params_from_gpt_params(params);
         guidance_ctx = llama_new_context_with_model(model, lparams);
     }
 
diff --git a/examples/perplexity/perplexity.cpp b/examples/perplexity/perplexity.cpp
index 768c2b400..fd4b03cb2 100644
--- a/examples/perplexity/perplexity.cpp
+++ b/examples/perplexity/perplexity.cpp
@@ -153,7 +153,7 @@ int main(int argc, char ** argv) {
     llama_context * ctx;
 
     // load the model and apply lora adapter, if any
-    std::tie(model, ctx, std::ignore) = llama_init_from_gpt_params(params);
+    std::tie(model, ctx) = llama_init_from_gpt_params(params);
     if (model == NULL) {
         fprintf(stderr, "%s: error: unable to load model\n", __func__);
         return 1;
diff --git a/examples/server/server.cpp b/examples/server/server.cpp
index 55cf1c94d..2cbfc0018 100644
--- a/examples/server/server.cpp
+++ b/examples/server/server.cpp
@@ -245,7 +245,7 @@ struct llama_server_context
     bool loadModel(const gpt_params &params_)
     {
         params = params_;
-        std::tie(model, ctx, std::ignore) = llama_init_from_gpt_params(params);
+        std::tie(model, ctx) = llama_init_from_gpt_params(params);
         if (model == nullptr)
         {
             LOG_ERROR("unable to load model", {{"model", params_.model}});
diff --git a/examples/simple/simple.cpp b/examples/simple/simple.cpp
index f59788865..2d913cebb 100644
--- a/examples/simple/simple.cpp
+++ b/examples/simple/simple.cpp
@@ -71,7 +71,7 @@ int main(int argc, char ** argv)
     llama_model * model;
     llama_context * ctx;
 
-    std::tie(model, ctx, std::ignore) = llama_init_from_gpt_params( params );
+    std::tie(model, ctx) = llama_init_from_gpt_params( params );
 
     if ( model == NULL )
     {