Upload folder using huggingface_hub

Test_RAG.py

@@ -290,18 +290,29 @@ if "GPU" in llm_device and "qwen2-7b-instruct" in llm_model_id:
 if llm_model_id == "red-pajama-3b-chat" and "GPU" in core.available_devices and llm_device in ["GPU", "AUTO"]:
     ov_config["INFERENCE_PRECISION_HINT"] = "f32"
 
-llm = HuggingFacePipeline.from_model_id(
-    model_id= "meta-llama/Meta-Llama-3-8B",                                                                        
-    #meta-llama/Meta-Llama-3-8B------------/meta-llama/Llama-3.2-3B-Instruct   
-    task="text-generation",
-    backend="openvino",
-    model_kwargs={
-        "device": llm_device,
-        "ov_config": ov_config,
-        "trust_remote_code": True,
-    },
-    pipeline_kwargs={"max_new_tokens": 2},
+# llm = HuggingFacePipeline.from_model_id(
+#     model_id= "meta-llama/Meta-Llama-3-8B",                                                                        
+#     #meta-llama/Meta-Llama-3-8B------------/meta-llama/Llama-3.2-3B-Instruct   
+#     task="text-generation",
+#     backend="openvino",
+#     model_kwargs={
+#         "device": llm_device,
+#         "ov_config": ov_config,
+#         "trust_remote_code": True,
+#     },
+#     pipeline_kwargs={"max_new_tokens": 2},
+# )
+from optimum.intel.openvino import OVModelForCausalLM
+
+llm = OVModelForCausalLM.from_pretrained(
+    model_id = "meta-llama/Meta-Llama-3-8B",
+    export=True,  # 将模型转换为 OpenVINO 格式
+    use_cache=False,
+    ov_config=ov_config,
+    trust_remote_code=True  # 支持远程代码的信任问题
 )
+
+
 # # 设置 pad_token_id 为 eos_token_id
 # from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 # tokenizer = AutoTokenizer.from_pretrained('meta-llama/Llama-3.2-3B-Instruct')