Update app.py

app.py

@@ -1,8 +1,20 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 
-# 🔁 CHANGE THIS to your own model name
-client = InferenceClient("SHIROI-07/skilllink-coach")
+# Use the base (untrained) model from Hugging Face Hub
+model_id = "mistralai/Mistral-7B-Instruct-v0.3"
+  # Your Hugging Face token
+
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id)
+
+pipe = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    max_new_tokens=512,
+    do_sample=True,
+)
 
 def respond(
     message,
@@ -12,29 +24,24 @@ def respond(
     temperature,
     top_p,
 ):
-    messages = [{"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]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    # 🧠 This uses the model like ChatGPT (if it's a chat-compatible model)
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    # Combine history and system message into a prompt
+    prompt = system_message.strip() + "\n"
+    for user, assistant in history:
+        if user:
+            prompt += f"User: {user}\n"
+        if assistant:
+            prompt += f"Assistant: {assistant}\n"
+    prompt += f"User: {message}\nAssistant:"
+
+    outputs = pipe(
+        prompt,
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-        response += token
-        yield response
+        pad_token_id=tokenizer.eos_token_id,
+    )
+    response = outputs[0]["generated_text"][len(prompt):]
+    yield response.strip()
 
 demo = gr.ChatInterface(
     respond,
@@ -42,9 +49,4 @@ demo = gr.ChatInterface(
         gr.Textbox(value="You are a professional AI coach helping people build skills.", 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=0.95, step=0.05, label="Top-p (nucleus sampling)"),
-    ],
-)
-
-if __name__ == "__main__":
-    demo.launch()
+        gr.Slider(minimum=0.1, maximum=1.0,