Update app.py

app.py

@@ -46,13 +46,13 @@ import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
 
 tokenizer = AutoTokenizer.from_pretrained("FlawedLLM/Bhashini_9")
-# quantization_config = BitsAndBytesConfig(
-#         load_in_4bit=True,
-#         bnb_4bit_use_double_quant=True,
-#         bnb_4bit_quant_type="nf4",  
-#         bnb_4bit_compute_dtype=torch.float16)
+quantization_config = BitsAndBytesConfig(
+        load_in_4bit=True,
+        bnb_4bit_use_double_quant=True,
+        bnb_4bit_quant_type="nf4",  
+        bnb_4bit_compute_dtype=torch.float16)
 torch_dtype =torch.float16
-model = AutoModelForCausalLM.from_pretrained("FlawedLLM/Bhashini_9",torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True,)
+model = AutoModelForCausalLM.from_pretrained("FlawedLLM/Bhashini_9",quantization_config=quantization_config ,)
 
 @spaces.GPU(duration=300)
 def chunk_it(input_command):