diff --git a/FlagEmbedding/abc/inference/__main__.py b/FlagEmbedding/abc/inference/__main__.py
deleted file mode 100644
index 80228c15..00000000
--- a/FlagEmbedding/abc/inference/__main__.py
+++ /dev/null
@@ -1,23 +0,0 @@
-from transformers import HfArgumentParser
-
-from FlagEmbedding.evaluation.beir import (
-    BEIREvalArgs, BEIREvalModelArgs,
-    BEIREvalRunner
-)
-
-
-parser = HfArgumentParser((
-    BEIREvalArgs,
-    BEIREvalModelArgs
-))
-
-eval_args, model_args = parser.parse_args_into_dataclasses()
-eval_args: BEIREvalArgs
-model_args: BEIREvalModelArgs
-
-runner = BEIREvalRunner(
-    eval_args=eval_args,
-    model_args=model_args
-)
-
-runner.run()
diff --git a/FlagEmbedding/inference/reranker/decoder_only/base.py b/FlagEmbedding/inference/reranker/decoder_only/base.py
index d16b87a9..7d0d1645 100644
--- a/FlagEmbedding/inference/reranker/decoder_only/base.py
+++ b/FlagEmbedding/inference/reranker/decoder_only/base.py
@@ -163,7 +163,6 @@ def __call__(self, data):
         return self.tokenizer.pad(
             data,
             padding=True,
-            max_length=self.max_len,
             pad_to_multiple_of=8,
             return_tensors='pt',
         )
diff --git a/FlagEmbedding/inference/reranker/decoder_only/lightweight.py b/FlagEmbedding/inference/reranker/decoder_only/lightweight.py
index b67892b1..87c0027d 100644
--- a/FlagEmbedding/inference/reranker/decoder_only/lightweight.py
+++ b/FlagEmbedding/inference/reranker/decoder_only/lightweight.py
@@ -81,7 +81,6 @@ def __call__(self, data):
         collected = self.tokenizer.pad(
             features,
             padding=True,
-            max_length=self.max_len,
             pad_to_multiple_of=8,
             return_tensors='pt',
         )
diff --git a/FlagEmbedding/inference/reranker/encoder_only/base.py b/FlagEmbedding/inference/reranker/encoder_only/base.py
index a6c47d2c..9a1abebe 100644
--- a/FlagEmbedding/inference/reranker/encoder_only/base.py
+++ b/FlagEmbedding/inference/reranker/encoder_only/base.py
@@ -178,7 +178,6 @@ def compute_score_single_gpu(
             inputs = self.tokenizer.pad(
                 sentences_batch,
                 padding=True,
-                max_length=max_length,
                 return_tensors='pt',
                 **kwargs
             ).to(device)