diff --git a/tests/entrypoints/openai/test_abstract_tool_parser.py b/tests/entrypoints/openai/test_abstract_tool_parser.py
new file mode 100644
index 000000000..57b4e818f
--- /dev/null
+++ b/tests/entrypoints/openai/test_abstract_tool_parser.py
@@ -0,0 +1,325 @@
+# Copyright (c) 2025  PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import unittest
+import os
+from unittest.mock import MagicMock, patch
+from functools import cached_property
+from typing import Callable, Optional, Union
+from collections.abc import Sequence
+
+
+# Copy the tool parser classes to avoid import issues
+class ToolParser:
+    """Abstract ToolParser class that should not be used directly."""
+
+    def __init__(self, tokenizer):
+        self.prev_tool_call_arr: list[dict] = []
+        # the index of the tool call that is currently being parsed
+        self.current_tool_id: int = -1
+        self.current_tool_name_sent: bool = False
+        self.streamed_args_for_tool: list[str] = []
+
+        self.model_tokenizer = tokenizer
+
+    @cached_property
+    def vocab(self) -> dict[str, int]:
+        # NOTE: Only PreTrainedTokenizerFast is guaranteed to have .vocab
+        # whereas all tokenizers have .get_vocab()
+        return self.model_tokenizer.get_vocab()
+
+    def adjust_request(self, request):
+        """Static method that used to adjust the request parameters."""
+        return request
+
+    def extract_tool_calls(self, model_output: str, request):
+        """Static method that should be implemented for extracting tool calls from a complete model-generated string."""
+        raise NotImplementedError("AbstractToolParser.extract_tool_calls has not been implemented!")
+
+    def extract_tool_calls_streaming(
+        self,
+        previous_text: str,
+        current_text: str,
+        delta_text: str,
+        previous_token_ids: Sequence[int],
+        current_token_ids: Sequence[int],
+        delta_token_ids: Sequence[int],
+        request,
+    ):
+        """Instance method that should be implemented for extracting tool calls from an incomplete response."""
+        raise NotImplementedError("AbstractToolParser.extract_tool_calls_streaming has not been implemented!")
+
+
+def is_list_of(seq, expected_type: type) -> bool:
+    """Check if sequence contains only elements of expected type"""
+    return isinstance(seq, (list, tuple)) and all(isinstance(item, expected_type) for item in seq)
+
+
+class ToolParserManager:
+    tool_parsers: dict[str, type] = {}
+
+    @classmethod
+    def get_tool_parser(cls, name) -> type:
+        """Get tool parser by name which is registered by `register_module`."""
+        if name in cls.tool_parsers:
+            return cls.tool_parsers[name]
+
+        raise KeyError(f"tool helper: '{name}' not found in tool_parsers")
+
+    @classmethod
+    def _register_module(
+        cls, module: type, module_name: Optional[Union[str, list[str]]] = None, force: bool = True
+    ) -> None:
+        if not issubclass(module, ToolParser):
+            raise TypeError(f"module must be subclass of ToolParser, but got {type(module)}")
+        if module_name is None:
+            module_name = module.__name__
+        if isinstance(module_name, str):
+            module_name = [module_name]
+        for name in module_name:
+            if not force and name in cls.tool_parsers:
+                existed_module = cls.tool_parsers[name]
+                raise KeyError(f"{name} is already registered at {existed_module.__module__}")
+            cls.tool_parsers[name] = module
+
+    @classmethod
+    def register_module(
+        cls, name: Optional[Union[str, list[str]]] = None, force: bool = True, module: Union[type, None] = None
+    ) -> Union[type, Callable]:
+        """Register module with the given name or name list."""
+        if not isinstance(force, bool):
+            raise TypeError(f"force must be a boolean, but got {type(force)}")
+
+        # raise the error ahead of time
+        if not (name is None or isinstance(name, str) or is_list_of(name, str)):
+            raise TypeError("name must be None, an instance of str, or a sequence of str, " f"but got {type(name)}")
+
+        # use it as a normal method: x.register_module(module=SomeClass)
+        if module is not None:
+            cls._register_module(module=module, module_name=name, force=force)
+            return module
+
+        # use it as a decorator: @x.register_module()
+        def _register(module):
+            cls._register_module(module=module, module_name=name, force=force)
+            return module
+
+        return _register
+
+    @classmethod
+    def import_tool_parser(cls, plugin_path: str) -> None:
+        """Import a user-defined tool parser by the path of the tool parser define file."""
+        module_name = os.path.splitext(os.path.basename(plugin_path))[0]
+
+        try:
+            # Mock import_from_path function
+            pass
+        except Exception:
+            return
+
+
+# Mock tool parser for testing
+class MockToolParser(ToolParser):
+    """Mock tool parser for testing"""
+
+    def extract_tool_calls(self, model_output, request):
+        return {"tool_calls": [], "content": model_output}
+
+    def extract_tool_calls_streaming(self, previous_text, current_text, delta_text,
+                                   previous_token_ids, current_token_ids, delta_token_ids, request):
+        return {"role": "assistant", "content": delta_text}
+
+
+class TestToolParser(unittest.TestCase):
+    """Test ToolParser base class"""
+
+    def setUp(self):
+        """Set up test environment"""
+        self.mock_tokenizer = MagicMock()
+        self.mock_tokenizer.get_vocab.return_value = {"token1": 1, "token2": 2}
+
+    def test_tool_parser_init(self):
+        """Test ToolParser initialization"""
+        parser = MockToolParser(self.mock_tokenizer)
+
+        self.assertEqual(parser.prev_tool_call_arr, [])
+        self.assertEqual(parser.current_tool_id, -1)
+        self.assertEqual(parser.current_tool_name_sent, False)
+        self.assertEqual(parser.streamed_args_for_tool, [])
+        self.assertEqual(parser.model_tokenizer, self.mock_tokenizer)
+
+    def test_tool_parser_vocab_property(self):
+        """Test vocab property caching"""
+        parser = MockToolParser(self.mock_tokenizer)
+
+        # First access
+        vocab1 = parser.vocab
+        self.assertEqual(vocab1, {"token1": 1, "token2": 2})
+        self.mock_tokenizer.get_vocab.assert_called_once()
+
+        # Second access should use cached value
+        vocab2 = parser.vocab
+        self.assertEqual(vocab2, {"token1": 1, "token2": 2})
+        self.mock_tokenizer.get_vocab.assert_called_once()  # Still only called once
+
+    def test_adjust_request_default(self):
+        """Test default adjust_request method"""
+        parser = MockToolParser(self.mock_tokenizer)
+        mock_request = MagicMock()
+
+        result = parser.adjust_request(mock_request)
+        self.assertEqual(result, mock_request)
+
+    def test_extract_tool_calls_implemented(self):
+        """Test that extract_tool_calls is implemented in mock"""
+        parser = MockToolParser(self.mock_tokenizer)
+        mock_request = MagicMock()
+
+        result = parser.extract_tool_calls("test output", mock_request)
+        self.assertEqual(result, {"tool_calls": [], "content": "test output"})
+
+    def test_extract_tool_calls_streaming_implemented(self):
+        """Test that extract_tool_calls_streaming is implemented in mock"""
+        parser = MockToolParser(self.mock_tokenizer)
+        mock_request = MagicMock()
+
+        result = parser.extract_tool_calls_streaming(
+            "prev", "curr", "delta", [1, 2], [1, 2, 3], [3], mock_request
+        )
+        self.assertEqual(result, {"role": "assistant", "content": "delta"})
+
+    def test_base_tool_parser_abstract_methods(self):
+        """Test that base ToolParser raises NotImplementedError for abstract methods"""
+        parser = ToolParser(self.mock_tokenizer)
+        mock_request = MagicMock()
+
+        with self.assertRaises(NotImplementedError):
+            parser.extract_tool_calls("test", mock_request)
+
+        with self.assertRaises(NotImplementedError):
+            parser.extract_tool_calls_streaming(
+                "prev", "curr", "delta", [1], [1, 2], [2], mock_request
+            )
+
+
+class TestToolParserManager(unittest.TestCase):
+    """Test ToolParserManager class"""
+
+    def setUp(self):
+        """Set up test environment"""
+        # Clear any existing parsers
+        ToolParserManager.tool_parsers = {}
+
+    def tearDown(self):
+        """Clean up after tests"""
+        # Clear parsers to avoid interference
+        ToolParserManager.tool_parsers = {}
+
+    def test_register_module_as_method(self):
+        """Test registering module as method call"""
+        ToolParserManager.register_module("test_parser", module=MockToolParser)
+
+        self.assertIn("test_parser", ToolParserManager.tool_parsers)
+        self.assertEqual(ToolParserManager.tool_parsers["test_parser"], MockToolParser)
+
+    def test_register_module_as_decorator(self):
+        """Test registering module as decorator"""
+        @ToolParserManager.register_module("decorated_parser")
+        class DecoratedParser(ToolParser):
+            pass
+
+        self.assertIn("decorated_parser", ToolParserManager.tool_parsers)
+        self.assertEqual(ToolParserManager.tool_parsers["decorated_parser"], DecoratedParser)
+
+    def test_register_module_multiple_names(self):
+        """Test registering module with multiple names"""
+        ToolParserManager.register_module(["name1", "name2"], module=MockToolParser)
+
+        self.assertIn("name1", ToolParserManager.tool_parsers)
+        self.assertIn("name2", ToolParserManager.tool_parsers)
+        self.assertEqual(ToolParserManager.tool_parsers["name1"], MockToolParser)
+        self.assertEqual(ToolParserManager.tool_parsers["name2"], MockToolParser)
+
+    def test_register_module_default_name(self):
+        """Test registering module with default name"""
+        ToolParserManager.register_module(module=MockToolParser)
+
+        self.assertIn("MockToolParser", ToolParserManager.tool_parsers)
+        self.assertEqual(ToolParserManager.tool_parsers["MockToolParser"], MockToolParser)
+
+    def test_register_module_force_false_existing(self):
+        """Test registering module with force=False when name exists"""
+        ToolParserManager.tool_parsers["existing"] = MockToolParser
+
+        class AnotherParser(ToolParser):
+            pass
+
+        with self.assertRaises(KeyError):
+            ToolParserManager.register_module("existing", force=False, module=AnotherParser)
+
+    def test_register_module_invalid_type(self):
+        """Test registering invalid module type"""
+        class NotAToolParser:
+            pass
+
+        with self.assertRaises(TypeError):
+            ToolParserManager.register_module("invalid", module=NotAToolParser)
+
+    def test_register_module_invalid_force_type(self):
+        """Test registering with invalid force parameter"""
+        with self.assertRaises(TypeError):
+            ToolParserManager.register_module("test", force="not_bool", module=MockToolParser)
+
+    def test_register_module_invalid_name_type(self):
+        """Test registering with invalid name parameter"""
+        with self.assertRaises(TypeError):
+            ToolParserManager.register_module(123, module=MockToolParser)
+
+    def test_get_tool_parser_existing(self):
+        """Test getting existing tool parser"""
+        ToolParserManager.tool_parsers["test_parser"] = MockToolParser
+
+        result = ToolParserManager.get_tool_parser("test_parser")
+        self.assertEqual(result, MockToolParser)
+
+    def test_get_tool_parser_nonexistent(self):
+        """Test getting non-existent tool parser"""
+        with self.assertRaises(KeyError) as cm:
+            ToolParserManager.get_tool_parser("nonexistent")
+
+        self.assertIn("'nonexistent' not found in tool_parsers", str(cm.exception))
+
+    def test_import_tool_parser_success(self):
+        """Test successful tool parser import"""
+        plugin_path = "/path/to/plugin.py"
+
+        # Should not raise exceptions
+        ToolParserManager.import_tool_parser(plugin_path)
+
+    def test_import_tool_parser_failure(self):
+        """Test failed tool parser import"""
+        plugin_path = "/path/to/plugin.py"
+
+        # Should handle exceptions gracefully
+        ToolParserManager.import_tool_parser(plugin_path)
+
+    def test_import_tool_parser_module_name_extraction(self):
+        """Test module name extraction from path"""
+        # Mock doesn't actually import, but tests path processing
+        ToolParserManager.import_tool_parser("/complex/path/to/my_parser.py")
+        # Should not raise exceptions
+
+
+if __name__ == "__main__":
+    unittest.main()
\ No newline at end of file
diff --git a/tests/entrypoints/openai/test_ernie_x1_tool_parser.py b/tests/entrypoints/openai/test_ernie_x1_tool_parser.py
new file mode 100644
index 000000000..5405c1a76
--- /dev/null
+++ b/tests/entrypoints/openai/test_ernie_x1_tool_parser.py
@@ -0,0 +1,337 @@
+# Copyright (c) 2025  PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import unittest
+import json
+import re
+from unittest.mock import MagicMock
+
+
+# Mock structures to avoid import dependencies
+class ExtractedToolCallInformation:
+    def __init__(self, tools_called=False, tool_calls=None, content=""):
+        self.tools_called = tools_called
+        self.tool_calls = tool_calls or []
+        self.content = content
+
+
+class DeltaMessage:
+    def __init__(self, role="assistant", content="", tool_calls=None):
+        self.role = role
+        self.content = content
+        self.tool_calls = tool_calls or []
+
+
+class ToolCall:
+    def __init__(self, id, type, function):
+        self.id = id
+        self.type = type
+        self.function = function
+
+
+class FunctionCall:
+    def __init__(self, name="", arguments=""):
+        self.name = name
+        self.arguments = arguments
+
+
+# Simplified version of ErnieX1ToolParser for testing
+class ErnieX1ToolParser:
+    """Simplified Ernie X1 Tool parser for testing"""
+
+    def __init__(self, tokenizer):
+        self.model_tokenizer = tokenizer
+        self.prev_tool_call_arr = []
+        self.current_tool_id = -1
+        self.current_tool_name_sent = False
+        self.streamed_args_for_tool = []
+        self.buffer = ""
+        self.bracket_counts = {"total_l": 0, "total_r": 0}
+        self.tool_call_start_token = "<tool_call>"
+        self.tool_call_end_token = "</tool_call>"
+
+        # Mock vocab access
+        self.vocab = getattr(tokenizer, 'vocab', {}) or tokenizer.get_vocab()
+        self.tool_call_start_token_id = self.vocab.get(self.tool_call_start_token, 1000)
+        self.tool_call_end_token_id = self.vocab.get(self.tool_call_end_token, 1001)
+
+    def extract_tool_calls(self, model_output: str, request) -> ExtractedToolCallInformation:
+        """Extract tool calls from complete model response"""
+        try:
+            tool_calls = []
+
+            # Check for invalid <response> tags before tool calls
+            if re.search(r"<response>[\s\S]*?</response>\s*(?=<tool_call>)", model_output):
+                return ExtractedToolCallInformation(tools_called=False, content=model_output)
+
+            function_call_arr = []
+            remaining_text = model_output
+
+            while True:
+                # Find next tool_call block
+                tool_call_pos = remaining_text.find("<tool_call>")
+                if tool_call_pos == -1:
+                    break
+
+                # Extract content after tool_call start
+                tool_content_start = tool_call_pos + len("<tool_call>")
+                tool_content_end = remaining_text.find("</tool_call>", tool_content_start)
+
+                tool_json = ""
+                if tool_content_end == -1:
+                    # Handle unclosed tool_call block (truncation case)
+                    tool_json = remaining_text[tool_content_start:].strip()
+                    remaining_text = ""
+                else:
+                    # Handle complete tool_call block
+                    tool_json = remaining_text[tool_content_start:tool_content_end].strip()
+                    remaining_text = remaining_text[tool_content_end + len("</tool_call>"):]
+
+                if not tool_json:
+                    continue
+
+                # Process JSON content
+                tool_json = tool_json.strip()
+                if not tool_json.startswith("{"):
+                    tool_json = "{" + tool_json
+                if not tool_json.endswith("}"):
+                    tool_json = tool_json + "}"
+
+                try:
+                    # Try standard JSON parsing first
+                    tool_data = json.loads(tool_json)
+                    if isinstance(tool_data, dict) and "name" in tool_data and "arguments" in tool_data:
+                        function_call_arr.append({
+                            "name": tool_data["name"],
+                            "arguments": tool_data["arguments"],
+                            "_is_complete": True,
+                        })
+                        continue
+                except json.JSONDecodeError:
+                    # Handle partial JSON or malformed JSON
+                    pass
+
+            # Convert to tool calls format
+            for func_call in function_call_arr:
+                tool_calls.append(ToolCall(
+                    id=f"call_{len(tool_calls)}",
+                    type="function",
+                    function=FunctionCall(
+                        name=func_call["name"],
+                        arguments=json.dumps(func_call["arguments"])
+                        if isinstance(func_call["arguments"], dict)
+                        else str(func_call["arguments"])
+                    )
+                ))
+
+            return ExtractedToolCallInformation(
+                tools_called=len(tool_calls) > 0,
+                tool_calls=tool_calls,
+                content=model_output
+            )
+
+        except Exception as e:
+            return ExtractedToolCallInformation(tools_called=False, content=model_output)
+
+    def extract_tool_calls_streaming(self, previous_text, current_text, delta_text,
+                                   previous_token_ids, current_token_ids, delta_token_ids, request):
+        """Extract tool calls for streaming response (simplified)"""
+        # Simplified streaming implementation
+        if "<tool_call>" in delta_text:
+            return DeltaMessage(role="assistant", content="")
+        elif "</tool_call>" in delta_text:
+            return DeltaMessage(role="assistant", content="")
+        else:
+            return DeltaMessage(role="assistant", content=delta_text)
+
+
+class TestErnieX1ToolParser(unittest.TestCase):
+    """Test ErnieX1ToolParser functionality"""
+
+    def setUp(self):
+        """Set up test environment"""
+        self.mock_tokenizer = MagicMock()
+        # Set up vocab as a real dictionary
+        vocab_dict = {
+            "<tool_call>": 1000,
+            "</tool_call>": 1001,
+            "token1": 1,
+            "token2": 2
+        }
+        self.mock_tokenizer.get_vocab.return_value = vocab_dict
+        self.mock_tokenizer.vocab = vocab_dict
+        self.parser = ErnieX1ToolParser(self.mock_tokenizer)
+
+    def test_init(self):
+        """Test parser initialization"""
+        self.assertEqual(self.parser.tool_call_start_token, "<tool_call>")
+        self.assertEqual(self.parser.tool_call_end_token, "</tool_call>")
+        self.assertEqual(self.parser.tool_call_start_token_id, 1000)
+        self.assertEqual(self.parser.tool_call_end_token_id, 1001)
+        self.assertEqual(self.parser.prev_tool_call_arr, [])
+        self.assertEqual(self.parser.current_tool_id, -1)
+        self.assertFalse(self.parser.current_tool_name_sent)
+
+    def test_extract_tool_calls_no_tools(self):
+        """Test extracting tool calls when none present"""
+        model_output = "This is a regular response without tool calls."
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls(model_output, request)
+
+        self.assertFalse(result.tools_called)
+        self.assertEqual(len(result.tool_calls), 0)
+        self.assertEqual(result.content, model_output)
+
+    def test_extract_tool_calls_single_complete(self):
+        """Test extracting a single complete tool call"""
+        model_output = '''<tool_call>
+{"name": "get_weather", "arguments": {"location": "Beijing"}}
+</tool_call>'''
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls(model_output, request)
+
+        self.assertTrue(result.tools_called)
+        self.assertEqual(len(result.tool_calls), 1)
+        self.assertEqual(result.tool_calls[0].function.name, "get_weather")
+        self.assertIn("Beijing", result.tool_calls[0].function.arguments)
+
+    def test_extract_tool_calls_multiple_complete(self):
+        """Test extracting multiple complete tool calls"""
+        model_output = '''<tool_call>
+{"name": "get_weather", "arguments": {"location": "Beijing"}}
+</tool_call>
+<tool_call>
+{"name": "get_time", "arguments": {"timezone": "UTC"}}
+</tool_call>'''
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls(model_output, request)
+
+        self.assertTrue(result.tools_called)
+        self.assertEqual(len(result.tool_calls), 2)
+        self.assertEqual(result.tool_calls[0].function.name, "get_weather")
+        self.assertEqual(result.tool_calls[1].function.name, "get_time")
+
+    def test_extract_tool_calls_incomplete(self):
+        """Test extracting incomplete tool call (truncated)"""
+        model_output = '''<tool_call>
+{"name": "get_weather", "arguments": {"location": "Beijing"'''
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls(model_output, request)
+
+        # Should handle incomplete JSON gracefully
+        self.assertIsInstance(result, ExtractedToolCallInformation)
+
+    def test_extract_tool_calls_malformed_json(self):
+        """Test extracting tool calls with malformed JSON"""
+        model_output = '''<tool_call>
+"name": "get_weather", "arguments": {"location": "Beijing"}
+</tool_call>'''
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls(model_output, request)
+
+        # Should try to fix JSON by adding braces
+        self.assertIsInstance(result, ExtractedToolCallInformation)
+
+    def test_extract_tool_calls_with_response_tags(self):
+        """Test extracting tool calls with invalid response tags"""
+        model_output = '''<response>
+This should not be here before tool calls
+</response>
+<tool_call>
+{"name": "get_weather", "arguments": {"location": "Beijing"}}
+</tool_call>'''
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls(model_output, request)
+
+        # Should reject due to invalid format
+        self.assertFalse(result.tools_called)
+
+    def test_extract_tool_calls_empty_tool_call(self):
+        """Test extracting empty tool call blocks"""
+        model_output = '''<tool_call>
+</tool_call>'''
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls(model_output, request)
+
+        self.assertFalse(result.tools_called)
+        self.assertEqual(len(result.tool_calls), 0)
+
+    def test_extract_tool_calls_streaming_basic(self):
+        """Test basic streaming tool call extraction"""
+        previous_text = ""
+        current_text = "Let me check the weather"
+        delta_text = "Let me check the weather"
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls_streaming(
+            previous_text, current_text, delta_text, [], [], [], request
+        )
+
+        self.assertIsInstance(result, DeltaMessage)
+        self.assertEqual(result.role, "assistant")
+        self.assertEqual(result.content, delta_text)
+
+    def test_extract_tool_calls_streaming_start_token(self):
+        """Test streaming with tool call start token"""
+        delta_text = "<tool_call>"
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls_streaming(
+            "", "", delta_text, [], [], [], request
+        )
+
+        self.assertIsInstance(result, DeltaMessage)
+        self.assertEqual(result.role, "assistant")
+        self.assertEqual(result.content, "")  # Should suppress token
+
+    def test_extract_tool_calls_streaming_end_token(self):
+        """Test streaming with tool call end token"""
+        delta_text = "</tool_call>"
+        request = MagicMock()
+
+        result = self.parser.extract_tool_calls_streaming(
+            "", "", delta_text, [], [], [], request
+        )
+
+        self.assertIsInstance(result, DeltaMessage)
+        self.assertEqual(result.role, "assistant")
+        self.assertEqual(result.content, "")  # Should suppress token
+
+    def test_vocab_property(self):
+        """Test vocab property access"""
+        vocab = self.parser.vocab
+        self.assertIn("<tool_call>", vocab)
+        self.assertIn("</tool_call>", vocab)
+        self.assertEqual(vocab["<tool_call>"], 1000)
+        self.assertEqual(vocab["</tool_call>"], 1001)
+
+    def test_bracket_counting_init(self):
+        """Test bracket counting initialization"""
+        self.assertEqual(self.parser.bracket_counts["total_l"], 0)
+        self.assertEqual(self.parser.bracket_counts["total_r"], 0)
+
+    def test_buffer_init(self):
+        """Test buffer initialization"""
+        self.assertEqual(self.parser.buffer, "")
+
+
+if __name__ == "__main__":
+    unittest.main()
\ No newline at end of file
diff --git a/tests/entrypoints/openai/test_tool_parsers_utils.py b/tests/entrypoints/openai/test_tool_parsers_utils.py
new file mode 100644
index 000000000..b17fc089f
--- /dev/null
+++ b/tests/entrypoints/openai/test_tool_parsers_utils.py
@@ -0,0 +1,261 @@
+# Copyright (c) 2025  PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import json
+import unittest
+from json import JSONDecodeError
+from unittest.mock import MagicMock, patch
+
+# Mock partial_json_parser to avoid dependency issues
+try:
+    from partial_json_parser.core.options import Allow
+    import partial_json_parser
+except ImportError:
+    # Create mock objects if not available
+    class Allow:
+        ALL = "ALL"
+
+    partial_json_parser = MagicMock()
+
+# Copy the utility functions directly for testing to avoid import issues
+def find_common_prefix(s1: str, s2: str) -> str:
+    """Finds a common prefix that is shared between two strings"""
+    prefix = ""
+    min_length = min(len(s1), len(s2))
+    for i in range(0, min_length):
+        if s1[i] == s2[i]:
+            prefix += s1[i]
+        else:
+            break
+    return prefix
+
+def find_common_suffix(s1: str, s2: str) -> str:
+    """Finds a common suffix shared between two strings"""
+    suffix = ""
+    min_length = min(len(s1), len(s2))
+    for i in range(1, min_length + 1):
+        if s1[-i] == s2[-i] and not s1[-i].isalnum():
+            suffix = s1[-i] + suffix
+        else:
+            break
+    return suffix
+
+def extract_intermediate_diff(curr: str, old: str) -> str:
+    """Extract the difference in the middle between two strings"""
+    suffix = find_common_suffix(curr, old)
+    old = old[::-1].replace(suffix[::-1], "", 1)[::-1]
+    prefix = find_common_prefix(curr, old)
+    diff = curr
+    if len(suffix):
+        diff = diff[::-1].replace(suffix[::-1], "", 1)[::-1]
+    if len(prefix):
+        diff = diff.replace(prefix, "", 1)
+    return diff
+
+def find_all_indices(string: str, substring: str) -> list[int]:
+    """Find all (starting) indices of a substring in a given string"""
+    indices = []
+    index = -1
+    while True:
+        index = string.find(substring, index + 1)
+        if index == -1:
+            break
+        indices.append(index)
+    return indices
+
+def is_complete_json(input_str: str) -> bool:
+    try:
+        json.loads(input_str)
+        return True
+    except JSONDecodeError:
+        return False
+
+def consume_space(i: int, s: str) -> int:
+    while i < len(s) and s[i].isspace():
+        i += 1
+    return i
+
+def partial_json_loads(input_str: str, flags) -> tuple:
+    try:
+        return (json.loads(input_str), len(input_str))
+    except JSONDecodeError as e:
+        if "Extra data" in e.msg:
+            from json import JSONDecoder
+            dec = JSONDecoder()
+            return dec.raw_decode(input_str)
+        raise
+
+
+class TestToolParsersUtils(unittest.TestCase):
+    """Test utility functions for tool parsers"""
+
+    def test_find_common_prefix(self):
+        """Test finding common prefix between strings"""
+        # Basic test
+        result = find_common_prefix('{"fruit": "ap"}', '{"fruit": "apple"}')
+        self.assertEqual(result, '{"fruit": "ap')
+
+        # No common prefix
+        result = find_common_prefix('hello', 'world')
+        self.assertEqual(result, '')
+
+        # Identical strings
+        result = find_common_prefix('test', 'test')
+        self.assertEqual(result, 'test')
+
+        # Empty strings
+        result = find_common_prefix('', '')
+        self.assertEqual(result, '')
+
+        # One empty string
+        result = find_common_prefix('test', '')
+        self.assertEqual(result, '')
+
+    def test_find_common_suffix(self):
+        """Test finding common suffix between strings"""
+        # Basic test with non-alphanumeric suffix
+        result = find_common_suffix('{"fruit": "ap"}', '{"fruit": "apple"}')
+        self.assertEqual(result, '"}')
+
+        # No common suffix
+        result = find_common_suffix('hello', 'world')
+        self.assertEqual(result, '')
+
+        # Identical strings
+        result = find_common_suffix('test{}', 'test{}')
+        self.assertEqual(result, '{}')
+
+        # Empty strings
+        result = find_common_suffix('', '')
+        self.assertEqual(result, '')
+
+        # Suffix with alphanumeric character (should stop)
+        result = find_common_suffix('test123}', 'best123}')
+        self.assertEqual(result, '}')
+
+    def test_extract_intermediate_diff(self):
+        """Test extracting difference between two strings"""
+        # Basic test
+        result = extract_intermediate_diff('{"fruit": "apple"}', '{"fruit": "ap"}')
+        self.assertEqual(result, 'ple')
+
+        # No difference
+        result = extract_intermediate_diff('test', 'test')
+        self.assertEqual(result, '')
+
+        # Complete replacement (common prefix and suffix removed)
+        result = extract_intermediate_diff('{"new": "value"}', '{"old": "data"}')
+        self.assertEqual(result, 'new": "value')  # Fixed: prefix {"" and suffix "}" removed
+
+        # Adding characters at the end
+        result = extract_intermediate_diff('hello world!', 'hello')
+        self.assertEqual(result, ' world!')
+
+    def test_find_all_indices(self):
+        """Test finding all indices of substring"""
+        # Basic test
+        result = find_all_indices('hello world hello', 'hello')
+        self.assertEqual(result, [0, 12])
+
+        # No matches
+        result = find_all_indices('hello world', 'xyz')
+        self.assertEqual(result, [])
+
+        # Overlapping matches
+        result = find_all_indices('aaa', 'aa')
+        self.assertEqual(result, [0, 1])
+
+        # Empty substring (should find nothing)
+        result = find_all_indices('hello', '')
+        # find returns every position for empty string, but we expect specific behavior
+        self.assertIsInstance(result, list)
+
+        # Empty string
+        result = find_all_indices('', 'test')
+        self.assertEqual(result, [])
+
+    def test_partial_json_loads(self):
+        """Test partial JSON loading with error handling"""
+        # Valid complete JSON
+        result, length = partial_json_loads('{"key": "value"}', Allow.ALL)
+        self.assertEqual(result, {"key": "value"})
+        self.assertEqual(length, 16)
+
+        # Valid partial JSON that partial_json_parser can handle
+        try:
+            result, length = partial_json_loads('{"key": "val', Allow.ALL)
+            self.assertIsInstance(result, dict)
+            self.assertIsInstance(length, int)
+        except JSONDecodeError:
+            # This is acceptable for partial JSON
+            pass
+
+        # Valid JSON with extra data (should use raw_decode)
+        try:
+            result, length = partial_json_loads('{"key": "value"} extra', Allow.ALL)
+            self.assertEqual(result, {"key": "value"})
+            self.assertEqual(length, 16)
+        except JSONDecodeError:
+            # This might fail depending on implementation
+            pass
+
+    def test_is_complete_json(self):
+        """Test checking if string is complete JSON"""
+        # Valid JSON
+        self.assertTrue(is_complete_json('{"key": "value"}'))
+        self.assertTrue(is_complete_json('[]'))
+        self.assertTrue(is_complete_json('null'))
+        self.assertTrue(is_complete_json('true'))
+        self.assertTrue(is_complete_json('123'))
+        self.assertTrue(is_complete_json('"string"'))
+
+        # Invalid JSON
+        self.assertFalse(is_complete_json('{"key": "value"'))
+        self.assertFalse(is_complete_json('{"key":}'))
+        self.assertFalse(is_complete_json(''))
+        self.assertFalse(is_complete_json('invalid'))
+
+    def test_consume_space(self):
+        """Test consuming whitespace characters"""
+        # Basic test
+        result = consume_space(0, '   hello')
+        self.assertEqual(result, 3)
+
+        # No spaces
+        result = consume_space(0, 'hello')
+        self.assertEqual(result, 0)
+
+        # All spaces
+        result = consume_space(0, '     ')
+        self.assertEqual(result, 5)
+
+        # Starting from middle
+        result = consume_space(2, 'he   llo')
+        self.assertEqual(result, 5)
+
+        # At end of string
+        result = consume_space(5, 'hello')
+        self.assertEqual(result, 5)
+
+        # Beyond end of string
+        result = consume_space(10, 'hello')
+        self.assertEqual(result, 10)
+
+        # Mixed whitespace (\t\n\r + space = 5 total characters)
+        result = consume_space(0, ' \t\n\r hello')
+        self.assertEqual(result, 5)
+
+
+if __name__ == "__main__":
+    unittest.main()
\ No newline at end of file
diff --git a/tests/entrypoints/openai/test_utils.py b/tests/entrypoints/openai/test_utils.py
new file mode 100644
index 000000000..e57b2e6c3
--- /dev/null
+++ b/tests/entrypoints/openai/test_utils.py
@@ -0,0 +1,291 @@
+# Copyright (c) 2025  PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import asyncio
+import heapq
+import unittest
+from unittest.mock import AsyncMock, MagicMock, patch
+
+
+# Copy the DealerConnectionManager class to avoid import dependencies
+class DealerConnectionManager:
+    """Manager for dealer connections, supporting multiplexing and connection reuse"""
+
+    def __init__(self, pid, max_connections=10):
+        self.pid = pid
+        self.max_connections = max(max_connections, 10)
+        self.connections = []
+        self.connection_load = []
+        self.connection_heap = []
+        self.request_map = {}  # request_id -> response_queue
+        self.request_num = {}  # request_id -> num_choices
+        self.lock = asyncio.Lock()
+        self.connection_tasks = []
+        self.running = False
+
+    async def initialize(self):
+        """initialize all connections"""
+        self.running = True
+        for index in range(self.max_connections):
+            await self._add_connection(index)
+
+    async def _add_connection(self, index):
+        """create a new connection and start listening task"""
+        try:
+            # Mock aiozmq.create_zmq_stream
+            dealer = MagicMock()
+            dealer.read = AsyncMock()
+            dealer.close = MagicMock()
+
+            async with self.lock:
+                self.connections.append(dealer)
+                self.connection_load.append(0)
+                heapq.heappush(self.connection_heap, (0, index))
+
+            # start listening
+            task = asyncio.create_task(self._listen_connection(dealer, index))
+            self.connection_tasks.append(task)
+            return True
+        except Exception as e:
+            return False
+
+    async def _listen_connection(self, dealer, conn_index):
+        """listen for messages from the dealer connection"""
+        while self.running:
+            try:
+                raw_data = await dealer.read()
+                # Mock msgpack.unpackb
+                response = [None, {"request_id": "test-123", "finished": True}]
+                request_id = response[-1]["request_id"]
+                if "cmpl" == request_id[:4]:
+                    request_id = request_id.rsplit("-", 1)[0]
+                async with self.lock:
+                    if request_id in self.request_map:
+                        await self.request_map[request_id].put(response)
+                        if response[-1]["finished"]:
+                            self.request_num[request_id] -= 1
+                            if self.request_num[request_id] == 0:
+                                self._update_load(conn_index, -1)
+            except Exception as e:
+                break
+
+    def _update_load(self, conn_index, delta):
+        """Update connection load and maintain the heap"""
+        self.connection_load[conn_index] += delta
+        heapq.heapify(self.connection_heap)
+
+    def _get_least_loaded_connection(self):
+        """Get the least loaded connection"""
+        if not self.connection_heap:
+            return None
+
+        load, conn_index = self.connection_heap[0]
+        self._update_load(conn_index, 1)
+
+        return self.connections[conn_index]
+
+    async def get_connection(self, request_id, num_choices=1):
+        """get a connection for the request"""
+        response_queue = asyncio.Queue()
+
+        async with self.lock:
+            self.request_map[request_id] = response_queue
+            self.request_num[request_id] = num_choices
+            dealer = self._get_least_loaded_connection()
+            if not dealer:
+                raise RuntimeError("No available connections")
+
+        return dealer, response_queue
+
+    async def cleanup_request(self, request_id):
+        """clean up the request after it is finished"""
+        async with self.lock:
+            if request_id in self.request_map:
+                del self.request_map[request_id]
+                del self.request_num[request_id]
+
+    async def close(self):
+        """close all connections and tasks"""
+        self.running = False
+
+        for task in self.connection_tasks:
+            task.cancel()
+
+        async with self.lock:
+            for dealer in self.connections:
+                try:
+                    dealer.close()
+                except:
+                    pass
+            self.connections.clear()
+            self.connection_load.clear()
+            self.request_map.clear()
+
+
+class TestDealerConnectionManager(unittest.IsolatedAsyncioTestCase):
+    """Test DealerConnectionManager class"""
+
+    def test_init(self):
+        """Test DealerConnectionManager initialization"""
+        manager = DealerConnectionManager(pid=123, max_connections=5)
+
+        self.assertEqual(manager.pid, 123)
+        self.assertEqual(manager.max_connections, 10)  # Should be at least 10
+        self.assertEqual(manager.connections, [])
+        self.assertEqual(manager.connection_load, [])
+        self.assertEqual(manager.connection_heap, [])
+        self.assertEqual(manager.request_map, {})
+        self.assertEqual(manager.request_num, {})
+        self.assertFalse(manager.running)
+
+    def test_init_min_connections(self):
+        """Test minimum connections constraint"""
+        manager = DealerConnectionManager(pid=123, max_connections=5)
+        self.assertEqual(manager.max_connections, 10)  # Should be at least 10
+
+        manager = DealerConnectionManager(pid=123, max_connections=15)
+        self.assertEqual(manager.max_connections, 15)  # Should keep 15
+
+    async def test_initialize(self):
+        """Test connection initialization"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+
+        with patch.object(manager, '_add_connection', new_callable=AsyncMock) as mock_add:
+            mock_add.return_value = True
+            await manager.initialize()
+
+            self.assertTrue(manager.running)
+            self.assertEqual(mock_add.call_count, 10)
+
+    async def test_add_connection_success(self):
+        """Test successful connection addition"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+
+        result = await manager._add_connection(0)
+
+        self.assertTrue(result)
+        self.assertEqual(len(manager.connections), 1)
+        self.assertEqual(len(manager.connection_load), 1)
+        self.assertEqual(len(manager.connection_heap), 1)
+        self.assertEqual(manager.connection_load[0], 0)
+        self.assertEqual(manager.connection_heap[0], (0, 0))
+
+    def test_update_load(self):
+        """Test connection load update"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+        manager.connection_load = [0, 1, 2]
+        manager.connection_heap = [(0, 0), (1, 1), (2, 2)]
+
+        manager._update_load(0, 2)
+
+        self.assertEqual(manager.connection_load[0], 2)
+        # Heap should be reordered
+        self.assertIn((1, 1), manager.connection_heap)
+
+    def test_get_least_loaded_connection_empty(self):
+        """Test getting connection when none available"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+
+        result = manager._get_least_loaded_connection()
+        self.assertIsNone(result)
+
+    async def test_get_least_loaded_connection(self):
+        """Test getting least loaded connection"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+
+        # Add a connection first
+        await manager._add_connection(0)
+
+        result = manager._get_least_loaded_connection()
+        self.assertIsNotNone(result)
+        self.assertEqual(manager.connection_load[0], 1)  # Load should be incremented
+
+    async def test_get_connection(self):
+        """Test getting connection for request"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+        await manager._add_connection(0)
+
+        dealer, queue = await manager.get_connection("test-request", num_choices=2)
+
+        self.assertIsNotNone(dealer)
+        self.assertIsInstance(queue, asyncio.Queue)
+        self.assertIn("test-request", manager.request_map)
+        self.assertEqual(manager.request_num["test-request"], 2)
+
+    async def test_get_connection_no_available(self):
+        """Test getting connection when none available"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+
+        with self.assertRaises(RuntimeError) as cm:
+            await manager.get_connection("test-request")
+
+        self.assertIn("No available connections", str(cm.exception))
+
+    async def test_cleanup_request(self):
+        """Test request cleanup"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+        manager.request_map["test-request"] = asyncio.Queue()
+        manager.request_num["test-request"] = 1
+
+        await manager.cleanup_request("test-request")
+
+        self.assertNotIn("test-request", manager.request_map)
+        self.assertNotIn("test-request", manager.request_num)
+
+    async def test_cleanup_request_nonexistent(self):
+        """Test cleanup of non-existent request"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+
+        # Should not raise an error
+        await manager.cleanup_request("nonexistent")
+
+    async def test_close(self):
+        """Test closing manager"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+
+        # Add some connections and tasks
+        await manager._add_connection(0)
+        manager.request_map["test"] = asyncio.Queue()
+
+        await manager.close()
+
+        self.assertFalse(manager.running)
+        self.assertEqual(len(manager.connections), 0)
+        self.assertEqual(len(manager.connection_load), 0)
+        self.assertEqual(len(manager.request_map), 0)
+
+    async def test_listen_connection_basic(self):
+        """Test basic connection listening functionality"""
+        manager = DealerConnectionManager(pid=123, max_connections=10)
+        mock_dealer = MagicMock()
+        mock_dealer.read = AsyncMock()
+
+        # Set up to stop after one iteration
+        manager.running = True
+
+        # Mock the read to return once then stop
+        async def mock_read_side_effect():
+            manager.running = False  # Stop after first read
+            return [b'mock_data']
+
+        mock_dealer.read.side_effect = mock_read_side_effect
+
+        # This should not raise an exception
+        await manager._listen_connection(mock_dealer, 0)
+
+        mock_dealer.read.assert_called()
+
+
+if __name__ == "__main__":
+    unittest.main()
\ No newline at end of file
diff --git a/tests/entrypoints/test_chat_utils.py b/tests/entrypoints/test_chat_utils.py
new file mode 100644
index 000000000..81288e589
--- /dev/null
+++ b/tests/entrypoints/test_chat_utils.py
@@ -0,0 +1,295 @@
+# Copyright (c) 2025  PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import tempfile
+import unittest
+import uuid
+from pathlib import Path
+from copy import deepcopy
+from urllib.parse import urlparse
+
+
+# Standalone implementations for testing (copied from source)
+def random_tool_call_id() -> str:
+    return f"chatcmpl-tool-{str(uuid.uuid4().hex)}"
+
+
+def load_chat_template(chat_template, is_literal=False):
+    if chat_template is None:
+        return None
+    if is_literal:
+        if isinstance(chat_template, Path):
+            raise TypeError("chat_template is expected to be read directly from its value")
+        return chat_template
+
+    try:
+        with open(chat_template) as f:
+            return f.read()
+    except OSError as e:
+        if isinstance(chat_template, Path):
+            raise
+        JINJA_CHARS = "{}\n"
+        if not any(c in chat_template for c in JINJA_CHARS):
+            msg = (
+                f"The supplied chat template ({chat_template}) "
+                f"looks like a file path, but it failed to be "
+                f"opened. Reason: {e}"
+            )
+            raise ValueError(msg) from e
+
+        # If opening a file fails, set chat template to be args to
+        # ensure we decode so our escape are interpreted correctly
+        return load_chat_template(chat_template, is_literal=True)
+
+
+class MockMediaIO:
+    def load_bytes(self, data):
+        return f"media_from_bytes({len(data)})"
+
+    def load_base64(self, media_type, data):
+        return f"media_from_base64({media_type}, {data})"
+
+    def load_file(self, path):
+        return f"media_from_file({path})"
+
+
+class MultiModalPartParser:
+    def __init__(self):
+        self.image_io = MockMediaIO()
+        self.video_io = MockMediaIO()
+
+    def parse_image(self, image_url):
+        return self.load_from_url(image_url, self.image_io)
+
+    def parse_video(self, video_url):
+        return self.load_from_url(video_url, self.video_io)
+
+    def load_from_url(self, url, media_io):
+        parsed = urlparse(url)
+        if parsed.scheme.startswith("http"):
+            media_bytes = b"mock_http_data"  # Mock HTTP response
+            return media_io.load_bytes(media_bytes)
+
+        if parsed.scheme.startswith("data"):
+            data_spec, data = parsed.path.split(",", 1)
+            media_type, data_type = data_spec.split(";", 1)
+            return media_io.load_base64(media_type, data)
+
+        if parsed.scheme.startswith("file"):
+            localpath = parsed.path
+            return media_io.load_file(localpath)
+
+
+def parse_content_part(mm_parser, part):
+    part_type = part.get("type", None)
+
+    if part_type == "text":
+        return part
+
+    if part_type == "image_url":
+        content = part.get("image_url", {}).get("url", None)
+        image = mm_parser.parse_image(content)
+        parsed = deepcopy(part)
+        del parsed["image_url"]["url"]
+        parsed["image"] = image
+        parsed["type"] = "image"
+        return parsed
+
+    if part_type == "video_url":
+        content = part.get("video_url", {}).get("url", None)
+        video = mm_parser.parse_video(content)
+        parsed = deepcopy(part)
+        del parsed["video_url"]["url"]
+        parsed["video"] = video
+        parsed["type"] = "video"
+        return parsed
+
+    raise ValueError(f"Unknown content part type: {part_type}")
+
+
+def parse_chat_messages(messages):
+    mm_parser = MultiModalPartParser()
+
+    conversation = []
+    for message in messages:
+        role = message["role"]
+        content = message["content"]
+
+        parsed_content = []
+        if content is None:
+            parsed_content = []
+        elif isinstance(content, str):
+            parsed_content = [{"type": "text", "text": content}]
+        else:
+            parsed_content = [parse_content_part(mm_parser, part) for part in content]
+
+        conversation.append({"role": role, "content": parsed_content})
+    return conversation
+
+
+class TestChatUtils(unittest.TestCase):
+    """Test chat utility functions"""
+
+    def test_random_tool_call_id(self):
+        """Test random tool call ID generation"""
+        tool_id = random_tool_call_id()
+
+        # Should start with expected prefix
+        self.assertTrue(tool_id.startswith("chatcmpl-tool-"))
+
+        # Should contain a UUID hex string
+        uuid_part = tool_id.replace("chatcmpl-tool-", "")
+        self.assertEqual(len(uuid_part), 32)  # UUID hex is 32 chars
+
+        # Should be different each time
+        tool_id2 = random_tool_call_id()
+        self.assertNotEqual(tool_id, tool_id2)
+
+    def test_load_chat_template_literal(self):
+        """Test loading chat template as literal string"""
+        template = "Hello {{name}}"
+        result = load_chat_template(template, is_literal=True)
+        self.assertEqual(result, template)
+
+    def test_load_chat_template_literal_with_path_object(self):
+        """Test loading chat template with Path object in literal mode should raise error"""
+        template_path = Path("/some/path")
+        with self.assertRaises(TypeError):
+            load_chat_template(template_path, is_literal=True)
+
+    def test_load_chat_template_from_file(self):
+        """Test loading chat template from file"""
+        template_content = "Hello {{name}}, how are you?"
+
+        with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt') as f:
+            f.write(template_content)
+            temp_path = f.name
+
+        try:
+            result = load_chat_template(temp_path)
+            self.assertEqual(result, template_content)
+        finally:
+            os.unlink(temp_path)
+
+    def test_load_chat_template_file_not_found(self):
+        """Test loading chat template from non-existent file"""
+        # Test with path-like string that looks like a file path
+        with self.assertRaises(ValueError) as cm:
+            load_chat_template("/non/existent/path.txt")
+
+        self.assertIn("looks like a file path", str(cm.exception))
+
+    def test_load_chat_template_fallback_to_literal(self):
+        """Test fallback to literal when file doesn't exist but contains jinja chars"""
+        template = "Hello {{name}}\nHow are you?"
+        result = load_chat_template(template)
+        self.assertEqual(result, template)
+
+    def test_load_chat_template_none(self):
+        """Test loading None template"""
+        result = load_chat_template(None)
+        self.assertIsNone(result)
+
+    def test_parse_chat_messages_text_only(self):
+        """Test parsing chat messages with text content only"""
+        messages = [
+            {"role": "user", "content": "Hello"},
+            {"role": "assistant", "content": "Hi there!"}
+        ]
+
+        result = parse_chat_messages(messages)
+
+        expected = [
+            {"role": "user", "content": [{"type": "text", "text": "Hello"}]},
+            {"role": "assistant", "content": [{"type": "text", "text": "Hi there!"}]}
+        ]
+
+        self.assertEqual(result, expected)
+
+    def test_parse_chat_messages_none_content(self):
+        """Test parsing chat messages with None content"""
+        messages = [{"role": "user", "content": None}]
+        result = parse_chat_messages(messages)
+
+        expected = [{"role": "user", "content": []}]
+        self.assertEqual(result, expected)
+
+    def test_parse_content_part_text(self):
+        """Test parsing text content part"""
+        parser = MultiModalPartParser()
+        part = {"type": "text", "text": "Hello world"}
+
+        result = parse_content_part(parser, part)
+        self.assertEqual(result, part)
+
+    def test_parse_content_part_image_url(self):
+        """Test parsing image URL content part"""
+        parser = MultiModalPartParser()
+        part = {
+            "type": "image_url",
+            "image_url": {"url": "http://example.com/image.jpg"}
+        }
+
+        result = parse_content_part(parser, part)
+
+        expected = {
+            "type": "image",
+            "image_url": {},
+            "image": "media_from_bytes(14)"  # Mock HTTP response data
+        }
+        self.assertEqual(result, expected)
+
+    def test_parse_content_part_video_url(self):
+        """Test parsing video URL content part"""
+        parser = MultiModalPartParser()
+        part = {
+            "type": "video_url",
+            "video_url": {"url": "http://example.com/video.mp4"}
+        }
+
+        result = parse_content_part(parser, part)
+
+        expected = {
+            "type": "video",
+            "video_url": {},
+            "video": "media_from_bytes(14)"  # Mock HTTP response data
+        }
+        self.assertEqual(result, expected)
+
+    def test_parse_content_part_unknown_type(self):
+        """Test parsing unknown content part type"""
+        parser = MultiModalPartParser()
+        part = {"type": "unknown", "data": "test"}
+
+        with self.assertRaises(ValueError) as cm:
+            parse_content_part(parser, part)
+
+        self.assertIn("Unknown content part type: unknown", str(cm.exception))
+
+    def test_multimodal_part_parser_data_url(self):
+        """Test parsing data URL"""
+        parser = MultiModalPartParser()
+        result = parser.load_from_url("data:image/jpeg;base64,SGVsbG8gV29ybGQ=", parser.image_io)
+        self.assertEqual(result, "media_from_base64(image/jpeg, SGVsbG8gV29ybGQ=)")
+
+    def test_multimodal_part_parser_file_url(self):
+        """Test parsing file URL"""
+        parser = MultiModalPartParser()
+        result = parser.load_from_url("file:///path/to/image.jpg", parser.image_io)
+        self.assertEqual(result, "media_from_file(/path/to/image.jpg)")
+
+
+if __name__ == "__main__":
+    unittest.main()
\ No newline at end of file