Updated to use token

app.py

@@ -18,8 +18,8 @@ def load_model_and_tokenizer(model_name):
     model = AutoModelForCausalLM.from_pretrained(model_name)
     return model, tokenizer
 
-model_8b, tokenizer_8b = load_model_and_tokenizer("meta-llama/Meta-Llama-3.1-8B")
-model_8b_instruct, tokenizer_8b_instruct = load_model_and_tokenizer("meta-llama/Meta-Llama-3.1-8B-Instruct")
+model_8b, tokenizer_8b = load_model_and_tokenizer("unsloth/Meta-Llama-3.1-8B-bnb-4bit")
+model_8b_instruct, tokenizer_8b_instruct = load_model_and_tokenizer("SanctumAI/Meta-Llama-3.1-8B-Instruct-GGUF")
 
 def generate_text(model, tokenizer, prompt, max_length=100):
     inputs = tokenizer(prompt, return_tensors="pt")
@@ -36,10 +36,10 @@ if st.button("Generate"):
     if prompt:
         col1, col2 = st.columns(2)
         
-        with col1:
-            st.subheader("LLaMA-3.1-8B Output")
-            output_8b = generate_text(model_8b, tokenizer_8b, prompt, max_length)
-            st.write(output_8b)
+        # with col1:
+        #     st.subheader("LLaMA-3.1-8B Output")
+        #     output_8b = generate_text(model_8b, tokenizer_8b, prompt, max_length)
+        #     st.write(output_8b)
         
         with col2:
             st.subheader("LLaMA-3.1-8B-Instruct Output")