Update app.py

app.py

@@ -1,25 +1,21 @@
-import gradio as gr
+import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import os
-import torch
+import gradio as gr
 
 model_name = "scb10x/llama-3-typhoon-v1.5x-70b-instruct-awq"
 token = os.getenv("HF_TOKEN")
 
-# Check if CUDA is available
-device = torch.device("cuda")
-torch.cuda.set_device(0)  # Use the first CUDA device
+# Remove these lines
+# device = torch.device("cuda")
+# torch.cuda.set_device(0)
 
 tokenizer = AutoTokenizer.from_pretrained(model_name, token=token)
-model = AutoModelForCausalLM.from_pretrained(model_name, token=token).to(device)
-
-print(f"CUDA available: {torch.cuda.is_available()}")
-print(f"Current device: {torch.cuda.current_device()}")
-print(f"Device name: {torch.cuda.get_device_name(0)}")
+model = AutoModelForCausalLM.from_pretrained(model_name, token=token)
 
 def generate_text(prompt):
-    inputs = tokenizer(prompt, return_tensors="pt").to(device)
+    inputs = tokenizer(prompt, return_tensors="pt")
     outputs = model.generate(inputs.input_ids, max_length=50)
     return tokenizer.decode(outputs[0], skip_special_tokens=True)
 
-gr.Interface(fn=generate_text, inputs="text", outputs="text").launch()
+gr.Interface(fn=generate_text, inputs="text", outputs="text").launch()