Update app.py

app.py

@@ -30,14 +30,15 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 model_id = "CohereForAI/c4ai-command-r-plus-4bit"
 
 
-tokenizer = AutoTokenizer.from_pretrained(model_id, token= token)
+tokenizer = AutoTokenizer.from_pretrained(model_id, token= token, quantization=False)
 
 model = AutoModelForCausalLM.from_pretrained(model_id, token= token, 
                                                  # torch_dtype=torch.bfloat16,    
                                              attn_implementation="flash_attention_2",
                                              # low_cpu_mem_usage=True,
                                              # llm_int8_enable_fp32_cpu_offload=True,
-                                             device_map="cuda"
+                                             device_map="cuda",
+                                             quantization=False
                                             )