klbr/khoj
1
0
Fork 0
mirror of https://github.com/khoaliber/khoj.git synced 2026-03-02 21:19:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

Put offline model response generation behind the chat lock as well

Browse Source 
Not just the chat response streaming
This commit is contained in:
Debanjum Singh Solanky
2023-08-01 21:52:24 -07:00
parent 1812473d27
commit 44292afff2
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
src/khoj/processor/conversation/gpt4all/chat_model.py
View File

@@ -165,8 +165,9 @@ def llm_thread(g, messages: List[ChatMessage], model: GPT4All):
templated_system_message = prompts.system_prompt_llamav2.format(message=system_message.content)
templated_user_message = prompts.general_conversation_llamav2.format(query=user_message.content)
prompted_message = templated_system_message + chat_history + templated_user_message
response_iterator = model.generate(prompted_message, streaming=True, max_tokens=500, n_batch=256)
state.chat_lock.acquire()
response_iterator = model.generate(prompted_message, streaming=True, max_tokens=500, n_batch=256)
try:
for response in response_iterator:
if any(stop_word in response.strip() for stop_word in stop_words):