Import turbomind in gradio server only when it is needed (#303)

InternLM · Aug 25, 2023 · 59f8e67 · 59f8e67
1 parent 4279d8c
commit 59f8e67
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/lmdeploy/serve/async_engine.py b/lmdeploy/serve/async_engine.py
@@ -6,7 +6,6 @@
 from contextlib import contextmanager
 from typing import Literal, Optional
 
-from lmdeploy import turbomind as tm
 from lmdeploy.model import MODELS, BaseModel
 
 
@@ -30,6 +29,7 @@ class AsyncEngine:
     """
 
     def __init__(self, model_path, instance_num=32, tp=1) -> None:
+        from lmdeploy import turbomind as tm
         from lmdeploy.turbomind.tokenizer import Tokenizer
         tokenizer_model_path = osp.join(model_path, 'triton_models',
                                         'tokenizer')