Merge branch 'frontier'

crazy_functions/理解PDF文档内容.py

@@ -19,7 +19,7 @@ def 解析PDF(file_name, llm_kwargs, plugin_kwargs, chatbot, history, system_pro
     TOKEN_LIMIT_PER_FRAGMENT = 2500
 
     from .crazy_utils import breakdown_txt_to_satisfy_token_limit_for_pdf
-    from request_llm.bridge_all import model_info
+    from request_llms.bridge_all import model_info
     enc = model_info["gpt-3.5-turbo"]['tokenizer']
     def get_token_num(txt): return len(enc.encode(txt, disallowed_special=()))
     paper_fragments = breakdown_txt_to_satisfy_token_limit_for_pdf(