Test

app.py

@@ -1,35 +1,22 @@
 import gradio as gr
 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
-"""
+# Use the CRA-v1-7B model (which uses the GGUF file internally)
 client = InferenceClient("molbal/CRA-v1-7B")
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [
-        {"role": "system", "content": "You are a writer’s assistant."},
-        {"role": "system", "content": system_message},
-    ]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
+def respond(message, history: list[tuple[str, str]], system_message, max_tokens, temperature, top_p):
+    # Build the conversation history; always include the system message
+    messages = [{"role": "system", "content": system_message}]
+    for user_msg, assistant_msg in history:
+        if user_msg:
+            messages.append({"role": "user", "content": user_msg})
+        if assistant_msg:
+            messages.append({"role": "assistant", "content": assistant_msg})
     messages.append({"role": "user", "content": message})
-
+    
     response = ""
-
-    for message in client.chat_completion(
+    # Call the model with streaming and the new parameters
+    for chunk in client.chat_completion(
         messages,
         max_tokens=max_tokens,
         stream=True,
@@ -38,31 +25,29 @@ def respond(
         num_ctx=16384,
         repeat_penalty=1.05,
     ):
-        token = message.choices[0].delta.content
-
+        token = chunk.choices[0].delta.content
         response += token
         yield response
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="Understand how the story flows, what motivations the characters have and how they will interact with each other and the world as a step by step thought process before continuing the story.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.8,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
-
-demo.launch()
-
-# Add an alert to mention that this runs on CPU
-gr.Markdown("**Note: This model runs on CPU, so it will be slow.**")
+# Create an alert message to inform users that inference runs on CPU (and will be slow)
+cpu_alert = gr.Markdown("**Note:** This model runs on CPU, so inference may be slow.")
+
+# Build the UI using Blocks to combine the alert and the ChatInterface
+with gr.Blocks() as demo:
+    cpu_alert.render()
+    chat_interface = gr.ChatInterface(
+        respond,
+        additional_inputs=[
+            gr.Textbox(
+                value="### System: You are a writer’s assistant. ### Task: Understand how the story flows, what motivations the characters have and how they will interact with each other and the world as a step by step thought process before continuing the story. ### Context: {context}",
+                label="System message"
+            ),
+            gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+            gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+            gr.Slider(minimum=0.1, maximum=1.0, value=0.8, step=0.05, label="Top-p (nucleus sampling)")
+        ]
+    )
+    chat_interface.render()
+
+if __name__ == "__main__":
+    demo.launch()