Skip to content

bugfix: reslove core dump of stream chat completion request when backend is VLM. #333

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 8, 2025
Merged

bugfix: reslove core dump of stream chat completion request when backend is VLM. #333

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 7 additions & 6 deletions xllm/api_service/chat_service_impl.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -344,7 +344,7 @@ bool send_delta_to_client_brpc(
std::shared_ptr<StreamOutputParser> stream_parser = nullptr) {
auto& response = call->response();

if (output.outputs.size() > 0) {
if (stream_parser && output.outputs.size() > 0) {
stream_parser->check_resize_for_index(output.outputs.size() - 1);
}
// send delta to client
Expand All @@ -371,7 +371,7 @@ bool send_delta_to_client_brpc(

// Handle reasoning text
if (!cur_text.empty()) {
if (stream_parser->is_reasoning()) {
if (stream_parser && stream_parser->is_reasoning()) {
auto parser = stream_parser->get_reasoning_parser(index);
auto result = parser->parse_stream_chunk(cur_text);
if (result.normal_text.has_value()) {
Expand All @@ -392,7 +392,7 @@ bool send_delta_to_client_brpc(

if (!cur_text.empty()) {
// Handle tool call text
if (stream_parser->is_tool_call()) {
if (stream_parser && stream_parser->is_tool_call()) {
if (!process_tool_call_stream(call,
stream_parser,
index,
Expand Down Expand Up @@ -422,7 +422,7 @@ bool send_delta_to_client_brpc(
// Handle finish reason
if (seq_output.finish_reason.has_value()) {
// Check for unstreamed tool args before sending finish reason
if (stream_parser->get_has_tool_call(index)) {
if (stream_parser && stream_parser->get_has_tool_call(index)) {
if (!check_for_unstreamed_tool_args(
call, stream_parser, index, request_id, created_time, model)) {
return false;
Expand All @@ -438,7 +438,7 @@ bool send_delta_to_client_brpc(
choice->set_index(index);
choice->mutable_delta();

if (stream_parser->get_has_tool_call(index) &&
if (stream_parser && stream_parser->get_has_tool_call(index) &&
seq_output.finish_reason.value() == "stop") {
choice->set_finish_reason("tool_calls");
} else {
Expand Down Expand Up @@ -615,7 +615,8 @@ void ChatServiceImpl::process_async_impl(std::shared_ptr<ChatCall> call) {
request_params.chat_template_kwargs, reasoning_parser_format_);

std::shared_ptr<StreamOutputParser> stream_parser;
if (request_params.streaming) {
if (request_params.streaming && (!tool_call_parser_format_.empty() ||
!reasoning_parser_format_.empty())) {
stream_parser =
std::make_shared<StreamOutputParser>(request_params.tools,
tool_call_parser_format_,
Expand Down