diff --git a/docker/docker-compose-base.yml b/docker/docker-compose-base.yml
index d2a142a82..98a4103bd 100644
--- a/docker/docker-compose-base.yml
+++ b/docker/docker-compose-base.yml
@@ -203,7 +203,7 @@ services:
     env_file: .env
     networks:
       - ragflow
-    command: ["--model-id", "/data/${TEI_MODEL}"]
+    command: ["--model-id", "/data/${TEI_MODEL}", "--auto-truncate"]
     restart: on-failure
 
 
@@ -217,7 +217,7 @@ services:
     env_file: .env
     networks:
       - ragflow
-    command: ["--model-id", "/data/${TEI_MODEL}"]
+    command: ["--model-id", "/data/${TEI_MODEL}", "--auto-truncate"]
     deploy:
       resources:
         reservations:
diff --git a/rag/llm/embedding_model.py b/rag/llm/embedding_model.py
index b433815b3..10eba69d3 100644
--- a/rag/llm/embedding_model.py
+++ b/rag/llm/embedding_model.py
@@ -82,7 +82,7 @@ class BuiltinEmbed(Base):
 
     def encode(self, texts: list):
         batch_size = 16
-        texts = [truncate(t, self._max_tokens) for t in texts]
+        # TEI is able to auto truncate inputs according to https://github.com/huggingface/text-embeddings-inference.
         token_count = 0
         ress = None
         for i in range(0, len(texts), batch_size):