[Bug]: Parsing error while patching Openai calls using litellm.completion with gpt-4o

[fenilfaldu]

Contact Details

[email protected]

📦 Package Version

0.3.26

🎞️ Framework Version

Litellm 1.58.2

🔎 Describe the Bug

@areibman At first it seemed it was an error from Litellm, but using Litellm independently it works fine.

Reference Script:

import os
import litellm
import agentops
agentops.init(default_tags=["Openai Example"])
messages = [{"role": "user", "content": "hello from litellm"}]
response = litellm.completion(
    model="gpt-4o", 
    messages=messages
)
print(response.choices[0].message.content)
agentops.end_session("Success") 

Traceback (most recent call last):
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\llms\openai\openai.py", line 429, in make_sync_openai_chat_completion_request       
    response = raw_response.parse()
AttributeError: 'str' object has no attribute 'parse'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\llms\openai\openai.py", line 691, in completion
    raise e
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\llms\openai\openai.py", line 619, in completion
    self.make_sync_openai_chat_completion_request(
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\llms\openai\openai.py", line 433, in make_sync_openai_chat_completion_request       
    raise Exception(
Exception: error - 'str' object has no attribute 'parse', Received response - <APIResponse [200 OK] type=<class 'openai.types.chat.chat_completion.ChatCompletion'>>, Type of response - <class 'str'>

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\main.py", line 1612, in completion
    raise e
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\main.py", line 1585, in completion
    response = openai_chat_completions.completion(
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\llms\openai\openai.py", line 701, in completion
    raise OpenAIError(
litellm.llms.openai.common_utils.OpenAIError: error - 'str' object has no attribute 'parse', Received response - <APIResponse [200 OK] type=<class 'openai.types.chat.chat_completion.ChatCompletion'>>, Type of response - <class 'str'>

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\Tracing.py", line 19, in <module>
    response = litellm.completion(
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\agentops\llms\providers\litellm.py", line 197, in patched_function
    result = self.original_create(*args, **kwargs)
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\utils.py", line 1030, in wrapper
    raise e
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\utils.py", line 906, in wrapper
    result = original_function(*args, **kwargs)
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\main.py", line 2967, in completion
    raise exception_type(
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\litellm_core_utils\exception_mapping_utils.py", line 2189, in exception_type        
    raise e
  File "C:\Users\LENOVO\Documents\AgentOps\agentops\.venv\lib\site-packages\litellm\litellm_core_utils\exception_mapping_utils.py", line 449, in exception_type
    raise APIError(
litellm.exceptions.APIError: litellm.APIError: APIError: OpenAIException - error - 'str' object has no attribute 'parse', Received response - <APIResponse [200 OK] type=<class 'openai.types.chat.chat_completion.ChatCompletion'>>, Type of response - <class 'str'>

🤝 Contribution

• Yes, I'd be happy to submit a pull request with these changes.
• I need some guidance on how to contribute.
• I'd prefer the AgentOps team to handle this update.