xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-30 17:18:43 +08:00
Code Issues Packages Projects Releases Wiki Activity

[gpt-oss] Fix harmony parser in streaming responses (#30205)

Browse Source 
Signed-off-by: AlonKejzman <alonkeizman@gmail.com>
This commit is contained in:
AlonKejzman 2025-12-22 14:56:06 +02:00 committed by GitHub
parent 256a33ecb4
commit bd6d5a7475
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 5 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
vllm/entrypoints/openai/serving_chat.py
View File

@ -811,6 +811,11 @@ class OpenAIServingChat(OpenAIServing):
delta_text += harmony_parser.last_content_delta or ""
cur_channel = harmony_parser.current_channel
cur_recipient = harmony_parser.current_recipient
# handle the case where several tokens where generated at once
# including the final token, leading to a delta in the text
# but the current channel to be empty (start state)
if not cur_channel and delta_text:
cur_channel = "final"
else:
delta_text = output.text