private-gpt
private-gpt copied to clipboard
Illegal instruction Mac M2
Can't run on mac m2, I get the following error:
Using embedded DuckDB with persistence: data will be stored in: db
Illegal instruction: 4
Environment (please complete the following information):
- OS / hardware: [e.g. macOS 13.3.1 / M2]
- Python version [e.g. 3.10.0]
.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
It works fine on my Mac with the M2 chip with the exact .env spec
macOS: 13.4
python 3.10.11
environment conda with all requirement dependencies installed.
Python 3.10.11 | packaged by conda-forge | (main, May 10 2023, 19:01:19) [Clang 14.0.6 ] on darwin
Type "help", "copyright", "credits" or "license" for more information.
Which model are you using? Seems to work for me with llamacpp but extremely slow.
I'm getting an error on my m1 too.
I'm getting an error on my m1 too.
what model are you using? For me it's currently working with llamacpp with the --mlock flag, which made it work much faster, I can even run 13b model now. But still getting the error with the gpt4all model in the readme.
Same error for me too on M1 mac and gpt4all model
Im getting a similar error:
zsh: Illegal Hardware Instruction
hope someone can help shed some light