diff --git a/openmanus_rl/__init__.py b/openmanus_rl/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/openmanus_rl/__pycache__/__init__.cpython-310.pyc b/openmanus_rl/__pycache__/__init__.cpython-310.pyc
new file mode 100644
index 00000000..9b17d985
Binary files /dev/null and b/openmanus_rl/__pycache__/__init__.cpython-310.pyc differ
diff --git a/openmanus_rl/__pycache__/configs.cpython-310.pyc b/openmanus_rl/__pycache__/configs.cpython-310.pyc
new file mode 100644
index 00000000..f8dce9d3
Binary files /dev/null and b/openmanus_rl/__pycache__/configs.cpython-310.pyc differ
diff --git a/openmanus_rl/__pycache__/sft.cpython-310.pyc b/openmanus_rl/__pycache__/sft.cpython-310.pyc
new file mode 100644
index 00000000..490c693e
Binary files /dev/null and b/openmanus_rl/__pycache__/sft.cpython-310.pyc differ
diff --git a/openmanus_rl/__pycache__/utils.cpython-310.pyc b/openmanus_rl/__pycache__/utils.cpython-310.pyc
new file mode 100644
index 00000000..5b14d5e3
Binary files /dev/null and b/openmanus_rl/__pycache__/utils.cpython-310.pyc differ
diff --git a/openmanus-rl/configs.py b/openmanus_rl/configs.py
similarity index 100%
rename from openmanus-rl/configs.py
rename to openmanus_rl/configs.py
diff --git a/openmanus-rl/grpo.py b/openmanus_rl/grpo.py
similarity index 100%
rename from openmanus-rl/grpo.py
rename to openmanus_rl/grpo.py
diff --git a/openmanus-rl/rewards.py b/openmanus_rl/rewards.py
similarity index 100%
rename from openmanus-rl/rewards.py
rename to openmanus_rl/rewards.py
diff --git a/openmanus-rl/sft.py b/openmanus_rl/sft.py
similarity index 100%
rename from openmanus-rl/sft.py
rename to openmanus_rl/sft.py
diff --git a/openmanus-rl/utils.py b/openmanus_rl/utils.py
similarity index 97%
rename from openmanus-rl/utils.py
rename to openmanus_rl/utils.py
index 1b99d586..59077727 100644
--- a/openmanus-rl/utils.py
+++ b/openmanus_rl/utils.py
@@ -3,7 +3,7 @@
 from transformers import AutoTokenizer, PreTrainedTokenizer
 from trl import ModelConfig
 
-from ..configs import GRPOConfig, SFTConfig
+from .configs import GRPOConfig, SFTConfig
 
 DEFAULT_CHAT_TEMPLATE = "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}"