AttributeError: 'GPT4All' object has no attribute 'model_type'

[45Matches]

Describe the bug and how to reproduce it
Following the setup instructions README.md. First time running the step python privateGPT.py after a successful ingest.py, the error is as follows:

% python privateGPT.py
Found model file.
gptj_model_load: loading model from 'models/ggml-gpt4all-j-v1.3-groovy.bin' - please wait ...
gptj_model_load: n_vocab = 50400
gptj_model_load: n_ctx   = 2048
gptj_model_load: n_embd  = 4096
gptj_model_load: n_head  = 16
gptj_model_load: n_layer = 28
gptj_model_load: n_rot   = 64
gptj_model_load: f16     = 2
gptj_model_load: ggml ctx size = 5401.45 MB
gptj_model_load: kv self size  =  896.00 MB
gptj_model_load: ................................... done
gptj_model_load: model size =  3609.38 MB / num tensors = 285
Traceback (most recent call last):
  File "/Users/user/.pyenv/versions/3.11.3/envs/vikings/privateGPT/privateGPT.py", line 76, in <module>
    main()
  File "/Users/user/.pyenv/versions/3.11.3/envs/vikings/privateGPT/privateGPT.py", line 36, in main
    llm = GPT4All(model=model_path, n_ctx=model_n_ctx, backend='gptj', callbacks=callbacks, verbose=False)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "pydantic/main.py", line 339, in pydantic.main.BaseModel.__init__
  File "pydantic/main.py", line 1102, in pydantic.main.validate_model
  File "/Users/user/.pyenv/versions/vikings/lib/python3.11/site-packages/langchain/llms/gpt4all.py", line 156, in validate_environment
    values["backend"] = values["client"].model_type
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'GPT4All' object has no attribute 'model_type'

Expected behavior
Running python privateGPT.py for the first time after successful installation, expecting to see the text > Enter your query

Environment

• MacOS 13.4 M1
• Python 3.11.3
• pyenv virtual

Additional context
.env

PERSIST_DIRECTORY=db
MODEL_TYPE=GPT4All
MODEL_PATH=models/ggml-gpt4all-j-v1.3-groovy.bin
EMBEDDINGS_MODEL_NAME=all-MiniLM-L6-v2
MODEL_N_CTX=1000
TARGET_SOURCE_CHUNKS=4

[beriagraphics]

Already fixed here #5657

[45Matches]

Thank you @beriagraphics. Nice one. I might be missing something, but the error is still occurring after #5657 was fixed? or is it a different model definition issue.

When I change directly model_type='gptj' in GPT4AllModel() init, the code works for me (i get the expected Enter a query: prompt, but not with the backend / client.model_type approach.

From Line 150 of gpt4all.py

        values["client"] = GPT4AllModel(
            model_name,
            model_path=model_path or None,
            model_type='gptj',
            allow_download=values["allow_download"],
        )
        # if values["n_threads"] is not None:
        #     # set n_threads
        #     values["client"].model.set_thread_count(values["n_threads"])
        # values["backend"] = values["client"].model_type

[45Matches]

Langchain Issue: #5720