Problem with GPT4All.___init___

[horvatm]

Hi.

Thank you for making py interface to GPT4All. I have tried

from pygpt4all import GPT4All
model = GPT4All('ggml-gpt4all-l13b-snoozy.bin')

with ggml-gpt4all-l13b-snoozy.bin having proper md5sum

md5sum ggml-gpt4all-l13b-snoozy.bin
91f886b68fbce697e9a3cd501951e455  ggml-gpt4all-l13b-snoozy.bin

but I get an error. Can you give a hint, what I am doing wrong?

The GPT4All_J works without a problem.

Best, Martin

---

TypeError Traceback (most recent call last)
Cell In[22], line 1
----> 1 model = GPT4All('ggml-gpt4all-l13b-snoozy.bin')

File ~/gpt4all-env/lib/python3.10/site-packages/pygpt4all/models/gpt4all.py:64, in GPT4All.init(self, model_path, prompt_context, prompt_prefix, prompt_suffix, log_level, n_ctx, seed, n_parts, f16_kv, logits_all, vocab_only, use_mlock, embedding)
62 # set logging level
63 set_log_level(log_level)
---> 64 super(GPT4All, self).init(model_path=model_path,
65 prompt_context=prompt_context,
66 prompt_prefix=prompt_prefix,
67 prompt_suffix=prompt_suffix,
68 log_level=log_level,
69 n_ctx=n_ctx,
70 seed=seed,
71 n_parts=n_parts,
72 f16_kv=f16_kv,
73 logits_all=logits_all,
74 vocab_only=vocab_only,
75 use_mlock=use_mlock,
76 embedding=embedding)

TypeError: Model.init() missing 1 required positional argument: 'ggml_model'

[abdeladim-s]

Hi @horvatm,

Thank you for using the interface.

if GPT4ALL_J is working then I think you are still using an old version of pyllamacpp.
please try ?

pip install -U pyllamacpp

[horvatm]

Yes, this did the trick. Thank you !!!