server : fix "model" in chat completion response

Author: ngxson

examples/server/utils.hpp

@@ -21,7 +21,7 @@
 #include <string>
 #include <vector>
 
-#define DEFAULT_OAICOMPAT_MODEL "gpt-3.5-turbo-0613"
+#define DEFAULT_OAICOMPAT_MODEL "llama-cpp"
 
 using json = nlohmann::ordered_json;
 using llama_tokens = std::vector<llama_token>;
@@ -590,6 +590,12 @@ static json oaicompat_completion_params_parse(
         throw std::runtime_error("Only one completion choice is allowed");
     }
 
+    // Handle "model" field
+    if (body.contains("model")) {
+        // this is used by format_*_response_oaicompat
+        llama_params["model"] = body.at("model").get<std::string>();
+    }
+
     // Handle "logprobs" field
     // TODO: The response format of this option is not yet OAI-compatible, but seems like no one really using it; We may need to fix it in the future
     if (json_value(body, "logprobs", false)) {