phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

Fix meta-reference GPU implementation for inference

Browse source
This commit is contained in:
Ashwin Bharambe 2025-01-22 18:31:59 -08:00
parent f4b0f2af8b
commit 23f1980f9c
2 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama_stack/providers/inline/inference/meta_reference/parallel_utils.py
View file

@ -357,8 +357,8 @@ class ModelParallelProcessGroup:
assert not self.running, "inference already running"
self.running = True
self.request_socket.send(encode_msg(TaskRequest(task=req)))
try:
self.request_socket.send(encode_msg(TaskRequest(task=req)))
while True:
obj_json = self.request_socket.recv()
obj = parse_message(obj_json)