Deprecate baichuan & internlm in favor of llama.cpp (#278)

README.md

@@ -23,6 +23,8 @@ Support Matrix:
 * Platforms: Linux, MacOS, Windows
 * Models: [ChatGLM-6B](https://github.com/THUDM/ChatGLM-6B), [ChatGLM2-6B](https://github.com/THUDM/ChatGLM2-6B), [ChatGLM3-6B](https://github.com/THUDM/ChatGLM3), [CodeGeeX2](https://github.com/THUDM/CodeGeeX2), [Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B), [Baichuan-7B](https://github.com/baichuan-inc/Baichuan-7B), [Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B), [Baichuan2](https://github.com/baichuan-inc/Baichuan2), [InternLM](https://github.com/InternLM/InternLM)
 
+**NOTE**: Baichuan & InternLM model series are deprecated in favor of [llama.cpp](https://github.com/ggerganov/llama.cpp).
+
 ## Getting Started
 
 **Preparation**

chatglm.cpp

@@ -1764,6 +1764,8 @@ Pipeline::Pipeline(const std::string &path, int max_length) {
         // load model
         model->load(loader);
     } else if (model_type == ModelType::BAICHUAN7B) {
+        std::cerr << "[WARN] Baichuan models are deprecated in favor of llama.cpp, and will be removed in next major "
+                     "version of chatglm.cpp\n";
         CHATGLM_CHECK(version == 1) << "only support version 1 for now but got " << version;
 
         // load config
@@ -1781,6 +1783,8 @@ Pipeline::Pipeline(const std::string &path, int max_length) {
         model = std::make_unique<Baichuan7BForCausalLM>(config);
         model->load(loader);
     } else if (model_type == ModelType::BAICHUAN13B) {
+        std::cerr << "[WARN] Baichuan models are deprecated in favor of llama.cpp, and will be removed in next major "
+                     "version of chatglm.cpp\n";
         CHATGLM_CHECK(version == 1) << "only support version 1 for now but got " << version;
 
         // load config
@@ -1798,6 +1802,8 @@ Pipeline::Pipeline(const std::string &path, int max_length) {
         model = std::make_unique<Baichuan13BForCausalLM>(config);
         model->load(loader);
     } else if (model_type == ModelType::INTERNLM) {
+        std::cerr << "[WARN] InternLM models are deprecated in favor of llama.cpp, and will be removed in next major "
+                     "version of chatglm.cpp\n";
         CHATGLM_CHECK(version == 1) << "only support version 1 for now but got " << version;
 
         // load config