.../venv/lib/python3.9/site-packages/urllib3/__init__.py:34: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020

# default python
python3 -m venv venv

# macports python3.11
python3.11 -m venv venv311

source ./venv/bin/activate

deactivate

# No Speed Limit, but lack of some packages
pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple

# Speed Limit to around 1MB/s 
pip config set global.index-url https://mirrors.aliyun.com/pypi/simple

# No Speed Limit
pip config set global.index-url https://mirror.baidu.com/pypi/simple

pip install --upgrade pip

pip install mlx
pip install huggingface_hub hf_transfer
pip install sentencepiece

pip install torch
pip install transformers
pip install sacremoses

pip install wheel xformers

export HF_ENDPOINT=https://huggingface.co/

# Enter venv

. ./venv/bin/activate 

git clone https://github.com/ml-explore/mlx-examples.git

# options to avoid re-download if already cached 
huggingface-cli download --cache-dir ~/.cache/huggingface/hub --local-dir Llama-2-7b-chat-mlx --local-dir-use-symlinks True  mlx-llama/Llama-2-7b-chat-mlx

python mlx-examples/llama/llama.py Llama-2-7b-chat-mlx/ Llama-2-7b-chat-mlx/tokenizer.model "What's the best Mac product ever"

(venv) sskaje@m3nimum apple-silicon-ai-notes % python mlx-examples/llama/llama.py Llama-2-7b-chat-mlx/ Llama-2-7b-chat-mlx/tokenizer.model "What's the best Mac product ever"

[INFO] Loading model from disk.
Press enter to start generation
------
?

There have been many great Mac products over the years, but some stand out as particularly memorable. Here are some of the best Mac products ever:

1. Macintosh 128k (1984) - This was the original Macintosh, released in 1984. It was the first commercially successful personal computer to use a graphical user interface (GUI), and it revolutionized the way people interacted with computers.
------
[INFO] Prompt processing: 0.348 s
[INFO] Full generation: 4.848 s

pip install llm 

# install llm mlx plugin
llm install https://github.com/sskaje/llm-mlx-llama/archive/refs/heads/main.zip
# use command below if github.com unreachable
#llm install https://github.com/sskaje/llm-mlx-llama/archive/refs/heads/main.zip
 

# download first 
# huggingface-cli download --cache-dir ~/.cache/huggingface/hub --local-dir Llama-2-7b-chat-mlx --local-dir-use-symlinks True  mlx-llama/Llama-2-7b-chat-mlx

llm -m mlx-llama \
  'five great reasons to get a pet pelican:' \
  -o model Llama-2-7b-chat-mlx \
  -o tokenizer Llama-2-7b-chat-mlx/tokenizer.model

python demo/marianmt_test.py

pip install xformers

git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui
cd stable-diffusion-webui

./webui.sh

python3 -m venv venv 
. venv/bin/activate

pip install pip --upgrade

pip install diffuzers
pip install altair==4.2.2

# this may report dependency error, but if you don't update, diffuzers won't work
pip install -U accelerate

diffuzers app

git clone https://github.com/lllyasviel/Fooocus.git
cd Fooocus

python3.11 -m venv venv
. venv/bin/activate

pip install --upgrade pip
pip install -r requirements_versions.txt

sed  -I "" -e 's/huggingface.co/huggingface.co/g' launch.py config_modification_tutorial.txt modules/config.py

python entry_with_update.py --disable-offload-from-vram