diff --git a/api/app/core/rag/app/naive.py b/api/app/core/rag/app/naive.py
index 72272347..93b96843 100644
--- a/api/app/core/rag/app/naive.py
+++ b/api/app/core/rag/app/naive.py
@@ -675,7 +675,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
             parser_config["chunk_token_num"] = 0
         else:
             sections = [(_, "") for _ in excel_parser(binary) if _]
-        parser_config["chunk_token_num"] = 12800
+        parser_config["chunk_token_num"] = 0
 
     elif re.search(r"\.(txt|py|js|java|c|cpp|h|php|go|ts|sh|cs|kt|sql)$", filename, re.IGNORECASE):
         callback(0.1, "Start to parse.")
diff --git a/api/app/core/rag/llm/embedding_model.py b/api/app/core/rag/llm/embedding_model.py
index 22e35a15..59210054 100644
--- a/api/app/core/rag/llm/embedding_model.py
+++ b/api/app/core/rag/llm/embedding_model.py
@@ -50,7 +50,9 @@ class OpenAIEmbed(Base):
     def encode(self, texts: list):
         # OpenAI requires batch size <=16
         batch_size = 16
-        texts = [truncate(t, 8191) for t in texts]
+        # Use 8000 instead of 8191 to leave safety margin for tokenizer differences
+        # between cl100k_base (used by truncate) and the actual embedding model
+        texts = [truncate(t, 8000) for t in texts]
         ress = []
         total_tokens = 0
         for i in range(0, len(texts), batch_size):
@@ -63,7 +65,7 @@ class OpenAIEmbed(Base):
         return np.array(ress), total_tokens
 
     def encode_queries(self, text):
-        res = self.client.embeddings.create(input=[truncate(text, 8191)], model=self.model_name, encoding_format="float",extra_body={"drop_params": True})
+        res = self.client.embeddings.create(input=[truncate(text, 8000)], model=self.model_name, encoding_format="float",extra_body={"drop_params": True})
         return np.array(res.data[0].embedding), self.total_token_count(res)
 
 
@@ -79,6 +81,7 @@ class LocalAIEmbed(Base):
 
     def encode(self, texts: list):
         batch_size = 16
+        texts = [truncate(t, 8000) for t in texts]
         ress = []
         for i in range(0, len(texts), batch_size):
             res = self.client.embeddings.create(input=texts[i : i + batch_size], model=self.model_name)
@@ -173,6 +176,7 @@ class XinferenceEmbed(Base):
 
     def encode(self, texts: list):
         batch_size = 16
+        texts = [truncate(t, 8000) for t in texts]
         ress = []
         total_tokens = 0
         for i in range(0, len(texts), batch_size):
@@ -188,7 +192,7 @@ class XinferenceEmbed(Base):
     def encode_queries(self, text):
         res = None
         try:
-            res = self.client.embeddings.create(input=[text], model=self.model_name)
+            res = self.client.embeddings.create(input=[truncate(text, 8000)], model=self.model_name)
             return np.array(res.data[0].embedding), self.total_token_count(res)
         except Exception as _e:
             log_exception(_e, res)
diff --git a/api/app/core/workflow/nodes/http_request/config.py b/api/app/core/workflow/nodes/http_request/config.py
index 3473f666..72474436 100644
--- a/api/app/core/workflow/nodes/http_request/config.py
+++ b/api/app/core/workflow/nodes/http_request/config.py
@@ -72,7 +72,8 @@ class HttpContentTypeConfig(BaseModel):
     @classmethod
     def validate_data(cls, v, info):
         content_type = info.data.get("content_type")
-        if content_type == HttpContentType.FROM_DATA and not isinstance(v, list):
+        if content_type == HttpContentType.FROM_DATA and (
+                not isinstance(v, list) or not all(isinstance(item, HttpFormData) for item in v)):
             raise ValueError("When content_type is 'form-data', data must be a list of HttpFormData")
         elif content_type in [HttpContentType.JSON] and not isinstance(v, str):
             raise ValueError("When content_type is JSON, data must be of type str")