Metadata-Version: 2.4
Name: applyllm
Version: 0.0.10
Summary: A python package to apply opensource LLM in local CUDA environment
Author-email: Yingding Wang <yingdingwang@yahoo.de>
Project-URL: Homepage, https://github.com/yingding/applyllm
Project-URL: Issues, https://github.com/yingding/applyllm/issues
Classifier: Programming Language :: Python :: 3
Classifier: Operating System :: OS Independent
Requires-Python: >=3.12
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: langchain>=1.2.15
Requires-Dist: transformers>=5.5.0
Requires-Dist: torch>=2.11.0
Requires-Dist: pydantic>=2.12.5
Requires-Dist: boto3>=1.42.9
Requires-Dist: mlflow>=3.7.0
Requires-Dist: bitsandbytes>=0.49.2
Requires-Dist: pypdf>=6.4.2
Requires-Dist: accelerate>=1.13.0
Requires-Dist: psycopg[binary]>=3.1.16
Requires-Dist: SQLAlchemy>=2.0.24
Requires-Dist: unstructured>=0.18.21
Requires-Dist: sentence-transformers>=5.2.0
Requires-Dist: docarray>=0.41.0
Dynamic: license-file

# applyllm: Training and Serving LLMs and SLMs

This repository contains the code for the [applyllm](https://pypi.org/project/applyllm/) python package, which supports loading and training open‑source LLM models such as LLaMA 2, Mixtral 8x7B, and etc. on CUDA and MPS.

## Install the package
```shell
pip install applyllm
```

