Update README.md

README.md

@@ -10,18 +10,9 @@ language:
 from transformers import AutoConfig, AutoTokenizer
 from optimum.intel.openvino import OVModelForCausalLM
 
-if model_to_run.value == "INT4":
-    model_dir = int4_model_dir
-elif model_to_run.value == "INT8":
-    model_dir = int8_model_dir
-else:
-    model_dir = fp16_model_dir
-print(f"Loading model from {model_dir}")
-
 ov_config = {"PERFORMANCE_HINT": "LATENCY", "NUM_STREAMS": "1", "CACHE_DIR": "", "INFERENCE_PRECISION_HINT": "f16"}
 
-model_name = model_configuration["model_id"]
-tok = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
+tok = AutoTokenizer.from_pretrained("xriminact/llama-3-8b-instruct-openvino", trust_remote_code=True)
 
 ov_model = OVModelForCausalLM.from_pretrained(
     "xriminact/llama-3-8b-instruct-openvino",