diff --git a/libs/community/langchain_community/llms/llamafile.py b/libs/community/langchain_community/llms/llamafile.py
index 5be6f4f2118..1aff521ee33 100644
--- a/libs/community/langchain_community/llms/llamafile.py
+++ b/libs/community/langchain_community/llms/llamafile.py
@@ -297,9 +297,10 @@ class Llamafile(LLM):
         for raw_chunk in response.iter_lines(decode_unicode=True):
             content = self._get_chunk_content(raw_chunk)
             chunk = GenerationChunk(text=content)
-            yield chunk
+
             if run_manager:
                 run_manager.on_llm_new_token(token=chunk.text)
+            yield chunk
 
     def _get_chunk_content(self, chunk: str) -> str:
         """When streaming is turned on, llamafile server returns lines like: