From aa95dc5568f9a27f17816426c3a8c118fc7789e3 Mon Sep 17 00:00:00 2001
From: Minsoo Cheong <icycle0409@snu.ac.kr>
Date: Tue, 5 Mar 2024 18:24:05 +0900
Subject: [PATCH] persist completion_id only for same stream

---
 examples/server/oai.hpp    | 4 ++--
 examples/server/server.cpp | 8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/examples/server/oai.hpp b/examples/server/oai.hpp
index 50ca48f93..5fe1a62a8 100644
--- a/examples/server/oai.hpp
+++ b/examples/server/oai.hpp
@@ -69,7 +69,7 @@ inline static json oaicompat_completion_params_parse(
     return llama_params;
 }
 
-inline static json format_final_response_oaicompat(const json &request, const task_result &response, std::string id, bool streaming = false)
+inline static json format_final_response_oaicompat(const json &request, const task_result &response, bool streaming = false)
 {
     json result = response.result_json;
 
@@ -105,7 +105,7 @@ inline static json format_final_response_oaicompat(const json &request, const ta
                 json{{"completion_tokens", num_tokens_predicted},
                      {"prompt_tokens",     num_prompt_tokens},
                      {"total_tokens",      num_tokens_predicted + num_prompt_tokens}}},
-            {"id", id}};
+            {"id", gen_chatcmplid()}};
 
     if (server_verbose) {
         res["__verbose"] = result;
diff --git a/examples/server/server.cpp b/examples/server/server.cpp
index 33d6faab6..f4a01397d 100644
--- a/examples/server/server.cpp
+++ b/examples/server/server.cpp
@@ -3210,8 +3210,7 @@ int main(int argc, char **argv)
                 res.set_content(models.dump(), "application/json; charset=utf-8");
             });
 
-    const std::string completion_id = gen_chatcmplid();
-    const auto chat_completions = [&llama, &validate_api_key, &sparams, &completion_id](const httplib::Request &req, httplib::Response &res)
+    const auto chat_completions = [&llama, &validate_api_key, &sparams](const httplib::Request &req, httplib::Response &res)
     {
         res.set_header("Access-Control-Allow-Origin", req.get_header_value("Origin"));
         if (!validate_api_key(req, res)) {
@@ -3228,7 +3227,7 @@ int main(int argc, char **argv)
             task_result result = llama.queue_results.recv(task_id);
 
             if (!result.error && result.stop) {
-                json oaicompat_result = format_final_response_oaicompat(data, result, completion_id);
+                json oaicompat_result = format_final_response_oaicompat(data, result);
 
                 res.set_content(oaicompat_result.dump(-1, ' ', false,
                                     json::error_handler_t::replace),
@@ -3239,7 +3238,8 @@ int main(int argc, char **argv)
             }
             llama.queue_results.remove_waiting_task_id(task_id);
         } else {
-            const auto chunked_content_provider = [task_id, &llama, &completion_id](size_t, httplib::DataSink &sink) {
+            const auto chunked_content_provider = [task_id, &llama](size_t, httplib::DataSink &sink) {
+                const std::string completion_id = gen_chatcmplid();
                 while (true) {
                     task_result llama_result = llama.queue_results.recv(task_id);
                     if (!llama_result.error) {