Update launch.py

launch.py

@@ -34,7 +34,7 @@ lm_model = LlamaForCausalLM.from_pretrained(
     torch_dtype=torch.float16,
     use_auth_token=AUTH_TOKEN,
     quantization_config=BitsAndBytesConfig(load_in_8bit=True))
-tokenizer = LlamaTokenizer.from_pretrained(LM_MODEL_PATH)
+tokenizer = LlamaTokenizer.from_pretrained(LM_MODEL_PATH, use_auth_token=AUTH_TOKEN)
 
 # visual model
 OPENAI_CLIP_MEAN = [0.48145466, 0.4578275, 0.40821073]
@@ -42,7 +42,7 @@ OPENAI_CLIP_STD = [0.26862954, 0.26130258, 0.27577711]
 # demo.py is in the project path, so we can use local path ".". Otherwise you should use "IDEA-CCNL/Ziya-BLIP2-14B-Visual-v1"
 model = AutoModelForCausalLM.from_pretrained(
     "IDEA-CCNL/Ziya-BLIP2-14B-Visual-v1", 
-    trust_remote_code=True, 
+    trust_remote_code=True, use_auth_token=AUTH_TOKEN,
     torch_dtype=torch.float16)
 model.cuda()  # if you use on cpu, comment this line
 model.language_model = lm_model