Update app.py

app.py

@@ -6,7 +6,7 @@ from transformers import pipeline
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
 #client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
+client = InferenceClient("distilbert/distilgpt2")
 modelpath = "distilgpt2"
 
 pipe = pipeline(
@@ -44,29 +44,19 @@ def respond(
     except Exception as e:
         return f"Error: {str(e)}"
 
-    try:
-        response = pipe(
-            messages,
-            max_new_tokens=max_tokens,
-            temperature=temperature,
-            top_p=top_p,
-        )[0]["generated_text"]
-        yield response
-    except Exception as e:
-        return f"Error: {str(e)}"
-    #response = ""
+    response = ""
 
-    #for message in client.chat_completion(
-    #    messages,
-    #    max_tokens=max_tokens,
-    #    stream=True,
-    #    temperature=temperature,
-    #    top_p=top_p,
-    #):
-    #    token = message.choices[0].delta.content
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+    ):
+        token = message.choices[0].delta.content
 
-    #    response += token
-    #    yield response
+        response += token
+        yield response
 
 
 """