pantry/projects/github.com/ggerganov/llama.cpp/package.yml
James Reynolds 2b06942c62
GitHub.com/ggerganov/llama.cpp update (#3696)
* llama.cpp, github version instead of hardcoded version
* llama.cpp, check if model is specified, if yes, run it, if not, then download model
* Use entrypoint for custom llama.cpp invocation
* `llama.cpp` is just raw executable. This I think is our new pattern.
* To run chat use the entrypoint: `pkgx +brewkit -- run llama.cpp`

Co-authored-by: James Reynolds <magnsuviri@me.com>
Co-authored-by: Max Howell <mxcl@me.com>
2023-10-26 07:24:04 -04:00

57 lines
1.3 KiB
YAML
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

distributable:
url: https://github.com/ggerganov/llama.cpp/archive/refs/tags/b{{version.raw}}.tar.gz
strip-components: 1
versions:
github: ggerganov/llama.cpp/tags
strip: /^b/
display-name:
LLaMA.cpp
provides:
- bin/llama.cpp
# NOTE! we do not “provide” convert.py. ∵ its too generic
# do `tea +github.comggerganovllama.cpp convert.py`
platforms:
- linux
- darwin/aarch64
# Illegal instruction: 4 on darwin/x86-64
dependencies:
python.org: ~3.11
pkgx.sh: ^1
build:
dependencies:
gnu.org/coreutils: '*'
env:
VIRTUAL_ENV: ${{prefix}}/venv
script:
- make --jobs {{hw.concurrency}}
- |
install -D main {{prefix}}/bin/llama.cpp
install -D props/entrypoint.sh {{prefix}}/entrypoint.sh
install -D ggml-metal.metal {{prefix}}/bin/ggml-metal.metal
- |
mkdir -p {{prefix}}/share
mv prompts {{prefix}}/share
- |
install -D convert.py $VIRTUAL_ENV/bin/convert.py
python-venv-stubber.sh convert.py
- |
python -m venv $VIRTUAL_ENV
source $VIRTUAL_ENV/bin/activate
pip install -r requirements.txt
deactivate
test:
'{{prefix}}/bin/llama.cpp --help'
# ^^ testing more than this requires downloading the models 😬
entrypoint: ./entrypoint.sh