Update app.py

app.py

@@ -1,20 +1,20 @@
 from transformers import AutoTokenizer, AutoModel
 import gradio as gr
 
+# chatglm-6b-int4 cuda，本地可以运行成功
 # tokenizer = AutoTokenizer.from_pretrained(".\\models\\chatglm-6b-int4", trust_remote_code=True, revision="")
-# model = AutoModel.from_pretrained(".\\models\\chatglm-6b-int4", trust_remote_code=True, revision="").half().cuda().cpu()
+# model = AutoModel.from_pretrained(".\\models\\chatglm-6b-int4", trust_remote_code=True, revision="").half().cuda()
 
 
-# from transformers import AutoTokenizer, AutoModel
-# tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True)
-# model = AutoModel.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True)
-# model = model.eval()
+# chatglm-6b CPU，
+tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True)
+model = AutoModel.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True).half()
 
 
+# chatglm-6b
 # kernel_file = "./models/chatglm-6b-int4/quantization_kernels.so"
-tokenizer = AutoTokenizer.from_pretrained("./models/chatglm-6b-int4", trust_remote_code=True, revision="")
+# tokenizer = AutoTokenizer.from_pretrained("./models/chatglm-6b-int4", trust_remote_code=True, revision="")
 # model = AutoModel.from_pretrained("./models/chatglm-6b-int4", trust_remote_code=True, revision="").half().cuda()
-model = AutoModel.from_pretrained("./models/chatglm-6b-int4", trust_remote_code=True, revision="").float()