diff --git a/common/common.cpp b/common/common.cpp
index d6a7ab753f6b3..4fd36105e42e4 100644
--- a/common/common.cpp
+++ b/common/common.cpp
@@ -846,7 +846,7 @@ struct common_init_result common_init_from_params(common_params & params) {
     } else if (!params.model_url.empty()) {
         model = common_load_model_from_url(params.model_url, params.model, params.hf_token, mparams);
     } else {
-        model = llama_load_model_from_file(params.model.c_str(), mparams);
+        model = llama_model_load_from_file(params.model.c_str(), mparams);
     }
 
     if (model == NULL) {
@@ -873,7 +873,7 @@ struct common_init_result common_init_from_params(common_params & params) {
         }
 
         if (!ok) {
-            llama_free_model(model);
+            llama_model_free(model);
 
             return iparams;
         }
@@ -884,7 +884,7 @@ struct common_init_result common_init_from_params(common_params & params) {
     llama_context * lctx = llama_new_context_with_model(model, cparams);
     if (lctx == NULL) {
         LOG_ERR("%s: failed to create context with model '%s'\n", __func__, params.model.c_str());
-        llama_free_model(model);
+        llama_model_free(model);
         return iparams;
     }
 
@@ -900,7 +900,7 @@ struct common_init_result common_init_from_params(common_params & params) {
         const auto cvec = common_control_vector_load(params.control_vectors);
         if (cvec.n_embd == -1) {
             llama_free(lctx);
-            llama_free_model(model);
+            llama_model_free(model);
 
             return iparams;
         }
@@ -913,7 +913,7 @@ struct common_init_result common_init_from_params(common_params & params) {
                                              params.control_vector_layer_end);
         if (err) {
             llama_free(lctx);
-            llama_free_model(model);
+            llama_model_free(model);
 
             return iparams;
         }
@@ -926,7 +926,7 @@ struct common_init_result common_init_from_params(common_params & params) {
         if (lora == nullptr) {
             LOG_ERR("%s: failed to apply lora adapter '%s'\n", __func__, la.path.c_str());
             llama_free(lctx);
-            llama_free_model(model);
+            llama_model_free(model);
             return iparams;
         }
 
@@ -1411,7 +1411,7 @@ struct llama_model * common_load_model_from_url(
         }
     }
 
-    return llama_load_model_from_file(local_path.c_str(), params);
+    return llama_model_load_from_file(local_path.c_str(), params);
 }
 
 struct llama_model * common_load_model_from_hf(
diff --git a/examples/batched-bench/batched-bench.cpp b/examples/batched-bench/batched-bench.cpp
index a3b21ad6bce44..dd75ff9f16319 100644
--- a/examples/batched-bench/batched-bench.cpp
+++ b/examples/batched-bench/batched-bench.cpp
@@ -38,7 +38,7 @@ int main(int argc, char ** argv) {
 
     llama_model_params model_params = common_model_params_to_llama(params);
 
-    llama_model * model = llama_load_model_from_file(params.model.c_str(), model_params);
+    llama_model * model = llama_model_load_from_file(params.model.c_str(), model_params);
 
     if (model == NULL) {
         fprintf(stderr , "%s: error: unable to load model\n" , __func__);
@@ -194,7 +194,7 @@ int main(int argc, char ** argv) {
     llama_batch_free(batch);
 
     llama_free(ctx);
-    llama_free_model(model);
+    llama_model_free(model);
 
     llama_backend_free();
 
diff --git a/examples/batched/batched.cpp b/examples/batched/batched.cpp
index 2e25b62f66b00..d34b030996ac2 100644
--- a/examples/batched/batched.cpp
+++ b/examples/batched/batched.cpp
@@ -41,7 +41,7 @@ int main(int argc, char ** argv) {
 
     llama_model_params model_params = common_model_params_to_llama(params);
 
-    llama_model * model = llama_load_model_from_file(params.model.c_str(), model_params);
+    llama_model * model = llama_model_load_from_file(params.model.c_str(), model_params);
 
     if (model == NULL) {
         LOG_ERR("%s: error: unable to load model\n" , __func__);
@@ -236,7 +236,7 @@ int main(int argc, char ** argv) {
 
     llama_sampler_free(smpl);
     llama_free(ctx);
-    llama_free_model(model);
+    llama_model_free(model);
 
     llama_backend_free();
 
diff --git a/examples/gritlm/gritlm.cpp b/examples/gritlm/gritlm.cpp
index 18a945b33905f..4d2db56249efc 100644
--- a/examples/gritlm/gritlm.cpp
+++ b/examples/gritlm/gritlm.cpp
@@ -165,7 +165,7 @@ int main(int argc, char * argv[]) {
 
     llama_backend_init();
 
-    llama_model * model = llama_load_model_from_file(params.model.c_str(), mparams);
+    llama_model * model = llama_model_load_from_file(params.model.c_str(), mparams);
 
     // create generation context
     llama_context * ctx = llama_new_context_with_model(model, cparams);
@@ -219,7 +219,7 @@ int main(int argc, char * argv[]) {
 
     llama_sampler_free(smpl);
     llama_free(ctx);
-    llama_free_model(model);
+    llama_model_free(model);
     llama_backend_free();
 
     return 0;
diff --git a/examples/llama-bench/llama-bench.cpp b/examples/llama-bench/llama-bench.cpp
index 2338ad1067dde..2a0916766099d 100644
--- a/examples/llama-bench/llama-bench.cpp
+++ b/examples/llama-bench/llama-bench.cpp
@@ -1526,10 +1526,10 @@ int main(int argc, char ** argv) {
         // keep the same model between tests when possible
         if (!lmodel || !prev_inst || !inst.equal_mparams(*prev_inst)) {
             if (lmodel) {
-                llama_free_model(lmodel);
+                llama_model_free(lmodel);
             }
 
-            lmodel = llama_load_model_from_file(inst.model.c_str(), inst.to_llama_mparams());
+            lmodel = llama_model_load_from_file(inst.model.c_str(), inst.to_llama_mparams());
             if (lmodel == NULL) {
                 fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, inst.model.c_str());
                 return 1;
@@ -1540,7 +1540,7 @@ int main(int argc, char ** argv) {
         llama_context * ctx = llama_new_context_with_model(lmodel, inst.to_llama_cparams());
         if (ctx == NULL) {
             fprintf(stderr, "%s: error: failed to create context with model '%s'\n", __func__, inst.model.c_str());
-            llama_free_model(lmodel);
+            llama_model_free(lmodel);
             return 1;
         }
 
@@ -1626,7 +1626,7 @@ int main(int argc, char ** argv) {
         ggml_threadpool_free_fn(threadpool);
     }
 
-    llama_free_model(lmodel);
+    llama_model_free(lmodel);
 
     if (p) {
         p->print_footer();
diff --git a/examples/llava/llava-cli.cpp b/examples/llava/llava-cli.cpp
index 2691c6e6b2dd2..27215a42e8a10 100644
--- a/examples/llava/llava-cli.cpp
+++ b/examples/llava/llava-cli.cpp
@@ -221,7 +221,7 @@ static struct llama_model * llava_init(common_params * params) {
 
     llama_model_params model_params = common_model_params_to_llama(*params);
 
-    llama_model * model = llama_load_model_from_file(params->model.c_str(), model_params);
+    llama_model * model = llama_model_load_from_file(params->model.c_str(), model_params);
     if (model == NULL) {
         LOG_ERR("%s: unable to load model\n" , __func__);
         return NULL;
@@ -265,7 +265,7 @@ static void llava_free(struct llava_context * ctx_llava) {
     }
 
     llama_free(ctx_llava->ctx_llama);
-    llama_free_model(ctx_llava->model);
+    llama_model_free(ctx_llava->model);
     llama_backend_free();
 }
 
@@ -323,7 +323,7 @@ int main(int argc, char ** argv) {
         }
     }
 
-    llama_free_model(model);
+    llama_model_free(model);
 
     return 0;
 }
diff --git a/examples/llava/minicpmv-cli.cpp b/examples/llava/minicpmv-cli.cpp
index e9cbb51ed90ab..2342bdd095642 100644
--- a/examples/llava/minicpmv-cli.cpp
+++ b/examples/llava/minicpmv-cli.cpp
@@ -31,7 +31,7 @@ static struct llama_model * llava_init(common_params * params) {
 
     llama_model_params model_params = common_model_params_to_llama(*params);
 
-    llama_model * model = llama_load_model_from_file(params->model.c_str(), model_params);
+    llama_model * model = llama_model_load_from_file(params->model.c_str(), model_params);
     if (model == NULL) {
         LOG_ERR("%s: unable to load model\n" , __func__);
         return NULL;
@@ -75,7 +75,7 @@ static void llava_free(struct llava_context * ctx_llava) {
     }
 
     llama_free(ctx_llava->ctx_llama);
-    llama_free_model(ctx_llava->model);
+    llama_model_free(ctx_llava->model);
     llama_backend_free();
 }
 
diff --git a/examples/llava/qwen2vl-cli.cpp b/examples/llava/qwen2vl-cli.cpp
index e86a60280aed6..f3e5d66e2c4e8 100644
--- a/examples/llava/qwen2vl-cli.cpp
+++ b/examples/llava/qwen2vl-cli.cpp
@@ -310,7 +310,7 @@ static struct llama_model * llava_init(common_params * params) {
 
     llama_model_params model_params = common_model_params_to_llama(*params);
 
-    llama_model * model = llama_load_model_from_file(params->model.c_str(), model_params);
+    llama_model * model = llama_model_load_from_file(params->model.c_str(), model_params);
     if (model == NULL) {
         LOG_ERR("%s: unable to load model\n" , __func__);
         return NULL;
@@ -354,7 +354,7 @@ static void llava_free(struct llava_context * ctx_llava) {
     }
 
     llama_free(ctx_llava->ctx_llama);
-    llama_free_model(ctx_llava->model);
+    llama_model_free(ctx_llava->model);
     llama_backend_free();
 }
 
@@ -575,7 +575,7 @@ int main(int argc, char ** argv) {
         }
     }
 
-    llama_free_model(model);
+    llama_model_free(model);
 
     return 0;
 }
diff --git a/examples/passkey/passkey.cpp b/examples/passkey/passkey.cpp
index 09bba708f6f91..ea91f376cd537 100644
--- a/examples/passkey/passkey.cpp
+++ b/examples/passkey/passkey.cpp
@@ -63,7 +63,7 @@ int main(int argc, char ** argv) {
 
     llama_model_params model_params = common_model_params_to_llama(params);
 
-    llama_model * model = llama_load_model_from_file(params.model.c_str(), model_params);
+    llama_model * model = llama_model_load_from_file(params.model.c_str(), model_params);
 
     if (model == NULL) {
         LOG_ERR("%s: unable to load model\n" , __func__);
@@ -266,7 +266,7 @@ int main(int argc, char ** argv) {
     llama_batch_free(batch);
 
     llama_free(ctx);
-    llama_free_model(model);
+    llama_model_free(model);
 
     llama_backend_free();
 
diff --git a/examples/quantize-stats/quantize-stats.cpp b/examples/quantize-stats/quantize-stats.cpp
index ab91d0b40aa03..9bfbb88623a16 100644
--- a/examples/quantize-stats/quantize-stats.cpp
+++ b/examples/quantize-stats/quantize-stats.cpp
@@ -309,7 +309,7 @@ int main(int argc, char ** argv) {
         auto mparams = llama_model_default_params();
         mparams.use_mlock  = false;
 
-        model = llama_load_model_from_file(params.model.c_str(), mparams);
+        model = llama_model_load_from_file(params.model.c_str(), mparams);
 
         if (model == NULL) {
             fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, params.model.c_str());
@@ -323,7 +323,7 @@ int main(int argc, char ** argv) {
 
         if (ctx == NULL) {
             fprintf(stderr, "%s: error: failed to create context with model '%s'\n", __func__, params.model.c_str());
-            llama_free_model(model);
+            llama_model_free(model);
             return 1;
         }
     }
@@ -347,7 +347,7 @@ int main(int argc, char ** argv) {
             fprintf(stderr, "%s: error: Quantization should be tested with a float model, "
                 "this model contains already quantized layers (%s is type %d)\n", __func__, kv_tensor.first.c_str(), kv_tensor.second->type);
             llama_free(ctx);
-            llama_free_model(model);
+            llama_model_free(model);
             return 1;
         }
         included_layers++;
@@ -409,7 +409,7 @@ int main(int argc, char ** argv) {
 
 
     llama_free(ctx);
-    llama_free_model(model);
+    llama_model_free(model);
     // report timing
     {
         const int64_t t_main_end_us = ggml_time_us();
diff --git a/examples/run/run.cpp b/examples/run/run.cpp
index 75b8172720238..c52a7961fb358 100644
--- a/examples/run/run.cpp
+++ b/examples/run/run.cpp
@@ -664,7 +664,7 @@ class LlamaData {
             "\r%*s"
             "\rLoading model",
             get_terminal_width(), " ");
-        llama_model_ptr model(llama_load_model_from_file(opt.model_.c_str(), opt.model_params));
+        llama_model_ptr model(llama_model_load_from_file(opt.model_.c_str(), opt.model_params));
         if (!model) {
             printe("%s: error: unable to load model from file: %s\n", __func__, opt.model_.c_str());
         }
diff --git a/examples/simple-chat/simple-chat.cpp b/examples/simple-chat/simple-chat.cpp
index 7f4da666b08ec..d72f5bcdde67c 100644
--- a/examples/simple-chat/simple-chat.cpp
+++ b/examples/simple-chat/simple-chat.cpp
@@ -69,7 +69,7 @@ int main(int argc, char ** argv) {
     llama_model_params model_params = llama_model_default_params();
     model_params.n_gpu_layers = ngl;
 
-    llama_model * model = llama_load_model_from_file(model_path.c_str(), model_params);
+    llama_model * model = llama_model_load_from_file(model_path.c_str(), model_params);
     if (!model) {
         fprintf(stderr , "%s: error: unable to load model\n" , __func__);
         return 1;
@@ -194,7 +194,7 @@ int main(int argc, char ** argv) {
     }
     llama_sampler_free(smpl);
     llama_free(ctx);
-    llama_free_model(model);
+    llama_model_free(model);
 
     return 0;
 }
diff --git a/examples/simple/simple.cpp b/examples/simple/simple.cpp
index 3288c0250a001..f691178904303 100644
--- a/examples/simple/simple.cpp
+++ b/examples/simple/simple.cpp
@@ -83,7 +83,7 @@ int main(int argc, char ** argv) {
     llama_model_params model_params = llama_model_default_params();
     model_params.n_gpu_layers = ngl;
 
-    llama_model * model = llama_load_model_from_file(model_path.c_str(), model_params);
+    llama_model * model = llama_model_load_from_file(model_path.c_str(), model_params);
 
     if (model == NULL) {
         fprintf(stderr , "%s: error: unable to load model\n" , __func__);
@@ -199,7 +199,7 @@ int main(int argc, char ** argv) {
 
     llama_sampler_free(smpl);
     llama_free(ctx);
-    llama_free_model(model);
+    llama_model_free(model);
 
     return 0;
 }
diff --git a/examples/tokenize/tokenize.cpp b/examples/tokenize/tokenize.cpp
index 57d9d43124184..684ca054aa487 100644
--- a/examples/tokenize/tokenize.cpp
+++ b/examples/tokenize/tokenize.cpp
@@ -338,7 +338,7 @@ int main(int raw_argc, char ** raw_argv) {
 
     llama_model_params model_params = llama_model_default_params();
     model_params.vocab_only = true;
-    llama_model * model = llama_load_model_from_file(model_path, model_params);
+    llama_model * model = llama_model_load_from_file(model_path, model_params);
     if (!model) {
         fprintf(stderr, "Error: could not load model from file '%s'.\n", model_path);
         return 1;
@@ -408,7 +408,7 @@ int main(int raw_argc, char ** raw_argv) {
     }
     // silence valgrind
     llama_free(ctx);
-    llama_free_model(model);
+    llama_model_free(model);
 
     return 0;
 }
diff --git a/include/llama-cpp.h b/include/llama-cpp.h
index 1500cb2fc9a7d..11306b17fac25 100644
--- a/include/llama-cpp.h
+++ b/include/llama-cpp.h
@@ -9,7 +9,7 @@
 #include "llama.h"
 
 struct llama_model_deleter {
-    void operator()(llama_model * model) { llama_free_model(model); }
+    void operator()(llama_model * model) { llama_model_free(model); }
 };
 
 struct llama_context_deleter {
diff --git a/include/llama.h b/include/llama.h
index 0f619aa19b7fd..0295a51fbee51 100644
--- a/include/llama.h
+++ b/include/llama.h
@@ -413,12 +413,19 @@ extern "C" {
     // Call once at the end of the program - currently only used for MPI
     LLAMA_API void llama_backend_free(void);
 
-    LLAMA_API struct llama_model * llama_load_model_from_file(
+    DEPRECATED(LLAMA_API struct llama_model * llama_load_model_from_file(
+                             const char * path_model,
+              struct llama_model_params   params),
+            "use llama_model_load_from_file instead");
+
+    LLAMA_API struct llama_model * llama_model_load_from_file(
                              const char * path_model,
               struct llama_model_params   params);
 
-    // TODO: rename to llama_model_free
-    LLAMA_API void llama_free_model(struct llama_model * model);
+    DEPRECATED(LLAMA_API void llama_free_model(struct llama_model * model),
+            "use llama_model_free instead");
+
+    LLAMA_API void llama_model_free(struct llama_model * model);
 
     // TODO: rename to llama_init_from_model
     LLAMA_API struct llama_context * llama_new_context_with_model(
diff --git a/src/llama-model.cpp b/src/llama-model.cpp
index 22596499a43ec..7deb3683bbccb 100644
--- a/src/llama-model.cpp
+++ b/src/llama-model.cpp
@@ -2009,6 +2009,10 @@ struct llama_model_params llama_model_default_params() {
 }
 
 void llama_free_model(struct llama_model * model) {
+    llama_model_free(model);
+}
+
+void llama_model_free(struct llama_model * model) {
     delete model;
 }
 
diff --git a/src/llama.cpp b/src/llama.cpp
index 4a6798f416fe9..7337c34ce573e 100644
--- a/src/llama.cpp
+++ b/src/llama.cpp
@@ -11656,6 +11656,12 @@ int64_t llama_time_us(void) {
 struct llama_model * llama_load_model_from_file(
         const char * path_model,
         struct llama_model_params params) {
+    return llama_model_load_from_file(path_model, params);
+}
+
+struct llama_model * llama_model_load_from_file(
+        const char * path_model,
+        struct llama_model_params params) {
     ggml_time_init();
 
     llama_model * model = new llama_model;
@@ -11694,7 +11700,7 @@ struct llama_model * llama_load_model_from_file(
         ggml_backend_reg_t rpc_reg = ggml_backend_reg_by_name("RPC");
         if (!rpc_reg) {
             LLAMA_LOG_ERROR("%s: failed to find RPC backend\n", __func__);
-            llama_free_model(model);
+            llama_model_free(model);
             return nullptr;
         }
 
@@ -11702,7 +11708,7 @@ struct llama_model * llama_load_model_from_file(
         ggml_backend_rpc_add_device_t ggml_backend_rpc_add_device_fn = (ggml_backend_rpc_add_device_t) ggml_backend_reg_get_proc_address(rpc_reg, "ggml_backend_rpc_add_device");
         if (!ggml_backend_rpc_add_device_fn) {
             LLAMA_LOG_ERROR("%s: failed to find RPC device add function\n", __func__);
-            llama_free_model(model);
+            llama_model_free(model);
             return nullptr;
         }
 
@@ -11712,7 +11718,7 @@ struct llama_model * llama_load_model_from_file(
                 model->devices.push_back(dev);
             } else {
                 LLAMA_LOG_ERROR("%s: failed to add RPC device for server '%s'\n", __func__, server.c_str());
-                llama_free_model(model);
+                llama_model_free(model);
                 return nullptr;
             }
         }
@@ -11744,7 +11750,7 @@ struct llama_model * llama_load_model_from_file(
     if (params.split_mode == LLAMA_SPLIT_MODE_NONE) {
         if (params.main_gpu < 0 || params.main_gpu >= (int)model->devices.size()) {
             LLAMA_LOG_ERROR("%s: invalid value for main_gpu: %d (available devices: %d)\n", __func__, params.main_gpu, (int)model->devices.size());
-            llama_free_model(model);
+            llama_model_free(model);
             return nullptr;
         }
         ggml_backend_dev_t main_gpu = model->devices[params.main_gpu];
@@ -11767,7 +11773,7 @@ struct llama_model * llama_load_model_from_file(
             LLAMA_LOG_INFO("%s: cancelled model load\n", __func__);
         }
 
-        llama_free_model(model);
+        llama_model_free(model);
         return nullptr;
     }
 
diff --git a/tests/test-autorelease.cpp b/tests/test-autorelease.cpp
index 57fa000114d5d..ba084a91a02ef 100644
--- a/tests/test-autorelease.cpp
+++ b/tests/test-autorelease.cpp
@@ -13,10 +13,10 @@ int main(int argc, char ** argv) {
 
     std::thread([&model_path]() {
         llama_backend_init();
-        auto * model = llama_load_model_from_file(model_path, llama_model_default_params());
+        auto * model = llama_model_load_from_file(model_path, llama_model_default_params());
         auto * ctx = llama_new_context_with_model(model, llama_context_default_params());
         llama_free(ctx);
-        llama_free_model(model);
+        llama_model_free(model);
         llama_backend_free();
     }).join();
 
diff --git a/tests/test-model-load-cancel.cpp b/tests/test-model-load-cancel.cpp
index 858535c3c4020..9095826fa9884 100644
--- a/tests/test-model-load-cancel.cpp
+++ b/tests/test-model-load-cancel.cpp
@@ -21,7 +21,7 @@ int main(int argc, char *argv[] ) {
         (void) ctx;
         return progress > 0.50;
     };
-    auto * model = llama_load_model_from_file(model_path, params);
+    auto * model = llama_model_load_from_file(model_path, params);
     llama_backend_free();
     return model == nullptr ? EXIT_SUCCESS : EXIT_FAILURE;
 }
diff --git a/tests/test-tokenizer-0.cpp b/tests/test-tokenizer-0.cpp
index 0af85f0020e19..121c2c60c9361 100644
--- a/tests/test-tokenizer-0.cpp
+++ b/tests/test-tokenizer-0.cpp
@@ -152,7 +152,7 @@ int main(int argc, char **argv) {
 
         mparams.vocab_only = true;
 
-        model = llama_load_model_from_file(fname.c_str(), mparams);
+        model = llama_model_load_from_file(fname.c_str(), mparams);
 
         if (model == NULL) {
             fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
@@ -165,7 +165,7 @@ int main(int argc, char **argv) {
 
         if (ctx == NULL) {
             fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
-            llama_free_model(model);
+            llama_model_free(model);
             return 1;
         }
     }
@@ -300,7 +300,7 @@ int main(int argc, char **argv) {
         fprintf(stderr, "%s : tokens written to '%s'\n", __func__, (fname_text + ".tokcpp").c_str());
     }
 
-    llama_free_model(model);
+    llama_model_free(model);
     llama_free(ctx);
 
     llama_backend_free();
diff --git a/tests/test-tokenizer-1-bpe.cpp b/tests/test-tokenizer-1-bpe.cpp
index 0ff7fc8333d8a..5718fab0401b8 100644
--- a/tests/test-tokenizer-1-bpe.cpp
+++ b/tests/test-tokenizer-1-bpe.cpp
@@ -46,7 +46,7 @@ int main(int argc, char **argv) {
 
         mparams.vocab_only = true;
 
-        model = llama_load_model_from_file(fname.c_str(), mparams);
+        model = llama_model_load_from_file(fname.c_str(), mparams);
 
         if (model == NULL) {
             fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
@@ -59,7 +59,7 @@ int main(int argc, char **argv) {
 
         if (ctx == NULL) {
             fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
-            llama_free_model(model);
+            llama_model_free(model);
             return 1;
         }
     }
@@ -143,7 +143,7 @@ int main(int argc, char **argv) {
         }
     }
 
-    llama_free_model(model);
+    llama_model_free(model);
     llama_free(ctx);
 
     llama_backend_free();
diff --git a/tests/test-tokenizer-1-spm.cpp b/tests/test-tokenizer-1-spm.cpp
index 9b0716a433332..ac05387c90949 100644
--- a/tests/test-tokenizer-1-spm.cpp
+++ b/tests/test-tokenizer-1-spm.cpp
@@ -34,7 +34,7 @@ int main(int argc, char ** argv) {
 
         mparams.vocab_only = true;
 
-        model = llama_load_model_from_file(fname.c_str(), mparams);
+        model = llama_model_load_from_file(fname.c_str(), mparams);
 
         if (model == NULL) {
             fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
@@ -47,7 +47,7 @@ int main(int argc, char ** argv) {
 
         if (ctx == NULL) {
             fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
-            llama_free_model(model);
+            llama_model_free(model);
             return 1;
         }
     }
@@ -113,7 +113,7 @@ int main(int argc, char ** argv) {
         }
     }
 
-    llama_free_model(model);
+    llama_model_free(model);
     llama_free(ctx);
 
     llama_backend_free();