Update app.py

app.py

@@ -1,79 +1,38 @@
-import torch
-import os
+import spaces
 import gradio as gr
-from huggingface_hub import InferenceClient, client
-# Use a pipeline as a high-level helper
+from transformers import AutoModelForCausalLM, AutoTokenizer
 from transformers import BitsAndBytesConfig
-from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
-HF_TOKEN = os.environ.get("HF_TOKEN", None)
 
 quantization_config = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_compute_dtype=torch.bfloat16)
-model_4bit = AutoModelForCausalLM.from_pretrained(
-    "unsloth/Llama-4-Scout-17B-16E-Instruct-unsloth-bnb-4bit",
-    quantization_config=quantization_config,
-    torch_dtype="auto"
-)
-# pipe = pipeline("image-text-to-text", model="")
-# pipe(messages)
-
-client = client(model_4bit)
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [
-    {"role": "user", "content": "Who are you?"},
-]
-    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 = ""
-
-    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
-
-
-"""
-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="You are a friendly Chatbot.", 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)",
-        ),
-    ],
-)
 
+model_name = "unsloth/Llama-4-Scout-17B-16E-Instruct-unsloth-bnb-4bit"
+
+@spaces.GPU(duration=180)
+def load_model():
+    model = AutoModelForCausalLM.from_pretrained(model_name, quantization_config=quantization_config)
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    return model, tokenizer
+
+@spaces.GPU
+def generate_text(prompt, model, tokenizer):
+    inputs = tokenizer(prompt, return_tensors="pt")
+    outputs = model.generate(**inputs, max_length=100)
+    return tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+def gradio_interface():
+    model, tokenizer = load_model()
+    
+    def wrapped_generate(prompt):
+        return generate_text(prompt, model, tokenizer)
+    
+    iface = gr.Interface(
+        fn=wrapped_generate,
+        inputs="text",
+        outputs="text",
+        title="Meta-Llama 4 Scout 17B Instruct 4bit bnb"
+    )
+    return iface
 
 if __name__ == "__main__":
-    demo.launch()
+    demo = gradio_interface()
+    demo.launch()