support batch infer in vllm

Former-commit-id: 3ef5ed3b9a44eed2f7e3ff221dfc343d0a97c0b5
2024-12-04 13:50:00 +00:00
parent 53edd62f8b
commit c1768cfb14
29 changed files with 148 additions and 407 deletions
--- a/src/llamafactory/webui/interface.py
+++ b/src/llamafactory/webui/interface.py
@@ -35,7 +35,7 @@ if is_gradio_available():

 def create_ui(demo_mode: bool = False) -> "gr.Blocks":
    engine = Engine(demo_mode=demo_mode, pure_chat=False)
-    hostname = os.getenv("HOSTNAME", os.getenv("COMPUTERNAME", platform.node())).split('.')[0]
+    hostname = os.getenv("HOSTNAME", os.getenv("COMPUTERNAME", platform.node())).split(".")[0]

    with gr.Blocks(title=f"LLaMA Board ({hostname})", css=CSS) as demo:
        if demo_mode: