From 7713e14d6a081674af331b0dfa4d2a9e09723800 Mon Sep 17 00:00:00 2001
From: Stephen Hu <stephenhu@seismic.com>
Date: Fri, 8 Aug 2025 14:13:07 +0800
Subject: [PATCH] Update chat_model.py (#9318)

### What problem does this PR solve?
https://github.com/infiniflow/ragflow/issues/9317
base on
https://discuss.ai.google.dev/t/valueerror-invalid-operation-the-response-text-quick-accessor-requires-the-response-to-contain-a-valid-part-but-none-were-returned/42866
should can be handled by retry
### Type of change

- [x] Refactoring
---
 rag/llm/chat_model.py | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/rag/llm/chat_model.py b/rag/llm/chat_model.py
index 49c535d81..1770b5b9e 100644
--- a/rag/llm/chat_model.py
+++ b/rag/llm/chat_model.py
@@ -1099,9 +1099,20 @@ class GeminiChat(Base):
 
         if system:
             self.model._system_instruction = content_types.to_content(system)
-        response = self.model.generate_content(hist, generation_config=gen_conf)
-        ans = response.text
-        return ans, response.usage_metadata.total_token_count
+        retry_count = 0
+        max_retries = 3
+        while retry_count < max_retries:
+            try:
+                response = self.model.generate_content(hist, generation_config=gen_conf)
+                ans = response.text
+                return ans, response.usage_metadata.total_token_count
+            except Exception as e:
+                retry_count += 1
+                if retry_count >= max_retries:
+                    raise e
+                else:
+                    import time
+                    time.sleep(50) 
 
     def chat_streamly(self, system, history, gen_conf={}, **kwargs):
         from google.generativeai.types import content_types