Update app.py

app.py

@@ -1,24 +1,17 @@
 import gradio as gr
-import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer, TextGenerationPipeline
+import requests
+import os
 
-MODEL_NAME = "deepseek-ai/deepseek-llm-7b-chat"  # Change for LLaMA if needed
+HF_API_KEY = os.getenv('TOKEN')
+MODEL_NAME = "deepseek-ai/deepseek-llm-7b-chat"
 
-# Load model & tokenizer
-tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
-model = AutoModelForCausalLM.from_pretrained(
-    MODEL_NAME,
-    torch_dtype=torch.float16,
-    device_map="auto"  # Uses available GPU/CPU
-)
-
-# Create text generation pipeline
-pipe = TextGenerationPipeline(model=model, tokenizer=tokenizer)
-
-# Define Gradio interface
 def chat(prompt):
-    output = pipe(prompt, max_length=512, do_sample=True, temperature=0.7)
-    return output[0]["generated_text"]
+    response = requests.post(
+        f"https://api-inference.huggingface.co/models/{MODEL_NAME}",
+        headers={"Authorization": f"Bearer {HF_API_KEY}"},
+        json={"inputs": prompt}
+    )
+    return response.json()[0]["generated_text"]
 
-iface = gr.Interface(fn=chat, inputs="text", outputs="text", title="DeepSeek 7B Chatbot")
+iface = gr.Interface(fn=chat, inputs="text", outputs="text", title="DeepSeek API Chatbot")
 iface.launch()