use text_generation instead of chat_completion, re-add top-k

app.py

@@ -4,34 +4,26 @@ from huggingface_hub import InferenceClient
 """
 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(moodel="pszemraj/distilgpt2-magicprompt-SD")
+client = InferenceClient(model="pszemraj/distilgpt2-magicprompt-SD")
 
 
 def respond(
     message,
-    history: list[tuple[str, str]],
+    _: list[tuple[str, str]],
     max_tokens: int,
     temperature: float,
-    top_p: float
+    top_p: float,
+    top_k: int
 ):
-    messages = []
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
     response = ""
 
-    for message in client.chat_completion(
-        messages,
+    for message in client.text_generation(
+        message,
         max_tokens=max_tokens,
         stream=True,
         temperature=temperature,
-        top_p=top_p
+        top_p=top_p,
+        top_k=top_k
     ):
         token = message.choices[0].delta.content
 
@@ -54,6 +46,13 @@ demo = gr.ChatInterface(
             step=0.05,
             label="Top-p",
         ),
+        gr.Slider(
+            minimum=10,
+            maximum=100,
+            value=30,
+            step=5,
+            label="Top-k",
+        ),
     ],
 )