tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-13 11:57:43 -04:00
Code Issues Packages Projects Releases Wiki Activity

server : do not get prompt in infill mode (#7286)

Browse Source 
* avoid to get prompt in infill mode and embedding mode

* remove embedding mode

* refactor format

---------

Co-authored-by: wudexiang <wudexiang@bytedance.com>
This commit is contained in:
woodx
2024-06-07 15:09:45 +08:00
committed by GitHub
parent d5c938cd77
commit a5cabd7649
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/server/server.cpp
View File

@@ -888,7 +888,7 @@ struct server_context {
slot.params.input_suffix = json_value(data, "input_suffix", default_params.input_suffix); slot.params.input_suffix = json_value(data, "input_suffix", default_params.input_suffix);
// get prompt // get prompt
{ if (!task.infill) {
const auto & prompt = data.find("prompt"); const auto & prompt = data.find("prompt");
if (prompt == data.end()) { if (prompt == data.end()) {
send_error(task, "Either \"prompt\" or \"messages\" must be provided", ERROR_TYPE_INVALID_REQUEST); send_error(task, "Either \"prompt\" or \"messages\" must be provided", ERROR_TYPE_INVALID_REQUEST);