Merge pull request #1520 from janhq/j/fix-streamlining-cortex-run

fix(#1425): Streamlining cortex run behavior

Author: namchuai

engine/cli/command_line_parser.cc

@@ -139,7 +139,7 @@ void CommandLineParser::SetupCommonCommands() {
   run_cmd->usage("Usage:\n" + commands::GetCortexBinary() +
                  " run [options] [model_id]");
   run_cmd->add_option("model_id", cml_data_.model_id, "");
-  run_cmd->add_flag("--chat", cml_data_.chat_flag, "Flag for interactive mode");
+  run_cmd->add_flag("-d,--detach", cml_data_.run_detach, "Detached mode");
   run_cmd->callback([this, run_cmd] {
     if (std::exchange(executed_, true))
       return;
@@ -151,7 +151,7 @@ void CommandLineParser::SetupCommonCommands() {
     commands::RunCmd rc(cml_data_.config.apiServerHost,
                         std::stoi(cml_data_.config.apiServerPort),
                         cml_data_.model_id, download_service_);
-    rc.Exec(cml_data_.chat_flag);
+    rc.Exec(cml_data_.run_detach);
   });
 
   auto chat_cmd = app_.add_subcommand(

engine/cli/command_line_parser.h

@@ -44,7 +44,7 @@ class CommandLineParser {
     std::string engine_src;
     std::string cortex_version;
     bool check_upd = true;
-    bool chat_flag = false;
+    bool run_detach = false;
     int port;
     config_yaml_utils::CortexConfig config;
     std::unordered_map<std::string, std::string> model_update_options;

engine/cli/commands/chat_cmd.cc

@@ -6,6 +6,6 @@ void ChatCmd::Exec(const std::string& host, int port,
                    const std::string& model_handle,
                    std::shared_ptr<DownloadService> download_service) {
   RunCmd rc(host, port, model_handle, download_service);
-  rc.Exec(true /*chat_flag*/);
+  rc.Exec(false /*detach mode*/);
 }
 };  // namespace commands

engine/cli/commands/run_cmd.cc

@@ -23,7 +23,7 @@ std::string Repo2Engine(const std::string& r) {
 };
 }  // namespace
 
-void RunCmd::Exec(bool chat_flag) {
+void RunCmd::Exec(bool run_detach) {
   std::optional<std::string> model_id = model_handle_;
 
   cortex::db::Models modellist_handler;
@@ -115,12 +115,12 @@ void RunCmd::Exec(bool chat_flag) {
     }
 
     // Chat
-    if (chat_flag) {
-      ChatCompletionCmd(model_service_).Exec(host_, port_, *model_id, mc, "");
-    } else {
+    if (run_detach) {
       CLI_LOG(*model_id << " model started successfully. Use `"
                         << commands::GetCortexBinary() << " chat " << *model_id
                         << "` for interactive chat shell");
+    } else {
+      ChatCompletionCmd(model_service_).Exec(host_, port_, *model_id, mc, "");
     }
   } catch (const std::exception& e) {
     CLI_LOG("Fail to run model with ID '" + model_handle_ + "': " + e.what());