Update app.py

app.py

@@ -8,7 +8,7 @@ from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
 st.set_page_config(page_title="AI Study Assistant", page_icon="🤖", layout="wide")
 
 # Set up the Groq API Key
-GROQ_API_KEY = "your_groq_api_key_here"  # Replace with your actual key
+GROQ_API_KEY = "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"  # Replace with your actual key
 os.environ["GROQ_API_KEY"] = GROQ_API_KEY
 
 # Initialize the Groq client
@@ -25,11 +25,13 @@ MODEL_NAME = "deepseek-ai/DeepSeek-R1"
 
 try:
     tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, trust_remote_code=True)
+    
     model = AutoModelForCausalLM.from_pretrained(
         MODEL_NAME,
         trust_remote_code=True,
-        torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
-        device_map="auto" if torch.cuda.is_available() else None
+        torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,  # ✅ Use FP16 on GPU, FP32 on CPU
+        device_map="auto" if torch.cuda.is_available() else None,  # ✅ Enable auto GPU usage
+        quantization_config=None  # ✅ Disable unsupported FP8 quantization
     )
 
     def generate_response_hf(user_message):