Litellm bug test

core/just_agents/base_agent.py

@@ -227,15 +227,6 @@ def streaming_query_with_current_memory(self, reconstruct_chunks = False, **kwar
                     else:
                         yield self._protocol.sse_wrap(part.model_dump(mode='json'))
 
-                # if self.tools and not self._tool_fuse_broken:
-                #     tool_calls = self._protocol.tool_calls_from_message(msg)
-                #     if tool_calls:
-                #         self.add_to_memory(
-                #             self._protocol.function_convention.reconstruct_tool_call_message(tool_calls)
-                #         )
-                #         self._process_function_calls(tool_calls)
-                #         tool_messages.append(self._process_function_calls(tool_calls))
-
             if len(self._partial_streaming_chunks) > 0:
                 assembly = self._protocol.response_from_deltas(self._partial_streaming_chunks)
                 self._partial_streaming_chunks.clear()

tests/test_pure_litellm.py

@@ -0,0 +1,64 @@
+import json
+import litellm
+import pytest
+from dotenv import load_dotenv
+
+
+
+def get_current_weather(location: str):
+    """
+    Gets the current weather in a given location
+    """
+    if "tokyo" in location.lower():
+        return json.dumps({"location": "Tokyo", "temperature": "10", "unit": "celsius"})
+    elif "san francisco" in location.lower():
+        return json.dumps({"location": "San Francisco", "temperature": "72", "unit": "fahrenheit"})
+    elif "paris" in location.lower():
+        return json.dumps({"location": "Paris", "temperature": "22", "unit": "celsius"})
+    else:
+        return json.dumps({"location": location, "temperature": "unknown"})
+
+@pytest.fixture(scope="module", autouse=True)
+def load_env():
+    load_dotenv(override=True)
+    messages = [
+        {"role": "system", "content": "You are a helpful AI assistant"},
+        {"role": "user", "content": "What's the weather like in San Francisco, Tokyo, and Paris?"}
+    ]
+    tools = [{"type": "function",
+              "function": litellm.utils.function_to_dict(get_current_weather)}]
+    OPENAI_GPT4oMINI = {
+        "messages": messages,
+        "model": "gpt-4o-mini",
+        "temperature": 0.0,
+        "tools": tools,
+        "stream": True
+    }
+    LLAMA3_3 = {
+        "messages" : messages,
+        "model": "groq/llama-3.3-70b-versatile",
+        "api_base": "https://api.groq.com/openai/v1",
+        "temperature": 0.0,
+        "tools": tools,
+        "stream" : True
+    }
+    return OPENAI_GPT4oMINI, LLAMA3_3
+
+def execute_completion(opts:dict):
+    partial_streaming_chunks = []
+    response_gen = litellm.completion(**opts)
+    for i, part in enumerate(response_gen):
+        partial_streaming_chunks.append(part)
+    assembly = litellm.stream_chunk_builder(partial_streaming_chunks)
+    print(assembly.choices[0].message.tool_calls)
+    assert len(assembly.choices[0].message.tool_calls) == 3, assembly.choices[0].message.tool_calls[0].function.arguments[0]
+    print (assembly.choices[0].message.tool_calls)
+
+def test_oai_works(load_env):
+    OPENAI_GPT4oMINI,_ = load_env
+    execute_completion(OPENAI_GPT4oMINI)
+
+@pytest.mark.skip(reason="until fixed in https://github.com/BerriAI/litellm/issues/7621")
+def test_grok_bug(load_env):
+    _, LLAMA3_3 = load_env
+    execute_completion(LLAMA3_3)