Remove quantization_config entirely to avoid bitsandbytes dependency

app.py

@@ -10,7 +10,7 @@ from transformers import BitsAndBytesConfig
 original_model = AutoModelForCausalLM.from_pretrained(
     "unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit",
     load_in_4bit=False,  # Disable 4-bit quantization
-    quantization_config=BitsAndBytesConfig()  # Use an empty quantization config
+    # Remove quantization_config entirely
 )
 original_tokenizer = AutoTokenizer.from_pretrained("unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit")
 
@@ -18,7 +18,7 @@ original_tokenizer = AutoTokenizer.from_pretrained("unsloth/DeepSeek-R1-Distill-
 fine_tuned_model = AutoModelForCausalLM.from_pretrained(
     "kas1/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit-John1",
     load_in_4bit=False,  # Disable 4-bit quantization
-    quantization_config=BitsAndBytesConfig()  # Use an empty quantization config
+    # Remove quantization_config entirely
 )
 fine_tuned_tokenizer = AutoTokenizer.from_pretrained("kas1/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit-John1")