Update app.py

app.py

@@ -8,31 +8,9 @@ import nncf
 
 import huggingface_hub as hf_hub
 # 初始化 OpenVINO 模型
-model_id = "Qwen/Qwen3-1.7B"
-model_path = "ov"
-'''
-import subprocess
-
-subprocess.run([
-    "optimum-cli", "export", "openvino",
-    "--model", "Qwen/Qwen3-1.7B",
-    #"--task", "text-generation-with-past",
-    "--weight-format", "int4",
-    "--quant-mode", "int4_f8e4m3",
-    "ov"
-])
-'''
-
-'''
 model_id = "OpenVINO/Qwen2.5-7B-Instruct-int4-ov"
 model_path = "ov"
-
-
-'''
-'''
 hf_hub.snapshot_download(model_id, local_dir=model_path)
-'''
-
 
 config = ov_genai.GenerationConfig()
 config.max_new_tokens = 4096
@@ -40,17 +18,6 @@ config.top_p = 0.9;
 config.top_k = 30;
 
 
-model = OVModelForCausalLM.from_pretrained(
-    model_id,
-    export=True,
-    quantization_config=OVWeightQuantizationConfig(bits=4),
-)
-
-
-# save the model after optimization
-model.save_pretrained(model_path)
-
-
 pipe = ov_genai.LLMPipeline(model_path, "CPU")
 
 #pipe.start_chat()