Update app.py

app.py

@@ -37,8 +37,8 @@ def terminate():
 
 import ollama
 import gradio as gr
-from ollama import Client
-client = Client(host='http://localhost:11434', timeout=120)
+from ollama import AsyncClient
+client = AsyncClient(host='http://localhost:11434', timeout=120)
 
 HF_TOKEN = os.environ.get("HF_TOKEN", None)
 
@@ -115,7 +115,7 @@ async def stream_chat(message: str, history: list, model: str, temperature: floa
         
     print(f"Conversation is -\n{conversation}")
                
-    response = client.chat(
+    async for part in await client.chat(
         model=model,
         stream=True,
         messages=conversation,
@@ -128,14 +128,8 @@ async def stream_chat(message: str, history: list, model: str, temperature: floa
             'repeat_penalty': penalty,
             'low_vram': True,
         },
-    )
-        
-    print(response)
-        
-    buffer = ""
-    for chunk in response:
-        buffer += chunk["message"]["content"]
-        yield buffer
+        ):
+            yield part['message']['content']
 
 
 async def main(message: str, history: list, model: str, temperature: float, max_new_tokens: int, top_p: float, top_k: int, penalty: float):
@@ -145,7 +139,7 @@ async def main(message: str, history: list, model: str, temperature: float, max_
     else:
         if not INIT_SIGN:
             yield "Please initialize Ollama"
-        else:         
+        else:
             async for response in stream_chat(
                 message,
                 history,