Pythagora Local LLM LLAMA 3 Ins 7B Q4 "There was a problem with request to openai API: string indices must be integers, not 'str'"

[ryanniccolls]

BASE::
MacBook Pro
13-inch, M1, 2020
Chip Apple M1 Memory 16 GB macOS Ventura 13.6.6
LM Studio
Version 0.2.21 (0.2.21)
Visual Studio Code Version: 188.1 Commit:

STEPS::
I changed .env to include local llm from lm studio

# OPENAI_ENDPOINT=https://api.openai.com/v1/chat/completions
OPENAI_ENDPOINT=http://localhost:8081/v1
OPENAI_API_KEY=lm-studio

In VS Studio I updated the location of the GPT -Pilot folder

BACKGROUND::
I can use the chat

ISSUE::
I cannot use the code creator

"There was a problem with request to openai API:
string indices must be integers, not 'str'"

"Do you want to try make the same request again? If yes, just press ENTER. Otherwise, type "no"."

NO::
Traceback (most recent call last):

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 219, in wrapper
  return func(*args, **kwargs)
         ^^^^^^^^^^^^^^^^^^^^^

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 502, in stream_gpt_completion
  raise ValueError(f'Error in LLM response: {json_line["error"]["message"]}')
                                             ~~~~~~~~~~~~~~~~~~^^^^^^^^^^^

TypeError: string indices must be integers, not 'str'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 151, in create_gpt_chat_completion
  response = stream_gpt_completion(gpt_data, req_type, project, model_name)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 301, in wrapper
  raise ApiError(f"Error making LLM API request: {err_str}") from e

helpers.exceptions.ApiError: Error making LLM API request: string indices must be integers, not 'str'

The request to OPENAI API failed with error: Error making LLM API request: string indices must be integers, not 'str'. Please try again later.
Error connecting to the API. Please check your API key/endpoint and try again.
Traceback (most recent call last):

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 219, in wrapper
  return func(*args, **kwargs)
         ^^^^^^^^^^^^^^^^^^^^^

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 502, in stream_gpt_completion
  raise ValueError(f'Error in LLM response: {json_line["error"]["message"]}')
                                             ~~~~~~~~~~~~~~~~~~^^^^^^^^^^^

TypeError: string indices must be integers, not 'str'


The above exception was the direct cause of the following exception:


Traceback (most recent call last):

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 81, in test_api_access
  response = create_gpt_chat_completion(messages, 'project_description', project)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 163, in create_gpt_chat_completion
  raise e

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 151, in create_gpt_chat_completion
  response = stream_gpt_completion(gpt_data, req_type, project, model_name)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Users/business/Documents/1 Projects/GPTProjects/PythagoraVSCodeExtension/gpt-pilot/pilot/utils/llm_connection.py", line 301, in wrapper
  raise ApiError(f"Error making LLM API request: {err_str}") from e

helpers.exceptions.ApiError: Error making LLM API request: string indices must be integers, not 'str'

POSSIBLE SOLUTION::
continuedev/continue#801
"Adding the "apiBase": "http://localhost:1234/v1/" line to the config fixes the issue, however it wasn't necessary a few days ago. Seems like a regression."

where is this config file?

Otherwise, I am in your hands.
Thank you