Metadata-Version: 2.4
Name: lighteval
Version: 0.10.1.dev0
Summary: A lightweight and configurable evaluation package
Author-email: Nathan Habib <nathan.habib@huggingface.com>, Clémentine Fourrier <clementine@huggingface.com>, Thomas Wolf <thom@huggingface.com>
Maintainer-email: Nathan Habib <nathan.habib@huggingface.com>, Clémentine Fourrier <clementine@huggingface.com>
License: MIT License
Project-URL: Homepage, https://github.com/huggingface/lighteval
Project-URL: Issues, https://github.com/huggingface/lighteval/issues
Keywords: evaluation,nlp,llm
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3.10
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: transformers>=4.51.0
Requires-Dist: accelerate
Requires-Dist: huggingface_hub[hf_xet]>=0.30.2
Requires-Dist: torch<3.0,>=2.0
Requires-Dist: GitPython>=3.1.41
Requires-Dist: datasets>=3.5.0
Requires-Dist: pydantic
Requires-Dist: numpy<2
Requires-Dist: typer
Requires-Dist: termcolor==2.3.0
Requires-Dist: pytablewriter
Requires-Dist: rich
Requires-Dist: colorlog
Requires-Dist: aenum==3.1.15
Requires-Dist: nltk==3.9.1
Requires-Dist: scikit-learn
Requires-Dist: sacrebleu
Requires-Dist: rouge_score==0.1.2
Requires-Dist: sentencepiece>=0.1.99
Requires-Dist: protobuf
Requires-Dist: pycountry
Requires-Dist: fsspec>=2023.12.2
Requires-Dist: httpx==0.27.2
Requires-Dist: latex2sympy2_extended==1.0.6
Provides-Extra: litellm
Requires-Dist: litellm; extra == "litellm"
Requires-Dist: diskcache; extra == "litellm"
Provides-Extra: tgi
Requires-Dist: text-generation>=0.6.0; extra == "tgi"
Provides-Extra: optimum
Requires-Dist: optimum==1.12.0; extra == "optimum"
Provides-Extra: quantization
Requires-Dist: bitsandbytes>=0.41.0; extra == "quantization"
Requires-Dist: auto-gptq>=0.4.2; extra == "quantization"
Provides-Extra: adapters
Requires-Dist: peft==0.3.0; extra == "adapters"
Provides-Extra: nanotron
Requires-Dist: nanotron; extra == "nanotron"
Requires-Dist: tensorboardX; extra == "nanotron"
Provides-Extra: tensorboardx
Requires-Dist: tensorboardX; extra == "tensorboardx"
Provides-Extra: vllm
Requires-Dist: vllm>=0.8.4; extra == "vllm"
Requires-Dist: ray; extra == "vllm"
Requires-Dist: more_itertools; extra == "vllm"
Provides-Extra: quality
Requires-Dist: ruff>=v0.11.0; extra == "quality"
Requires-Dist: pre-commit; extra == "quality"
Provides-Extra: tests
Requires-Dist: pytest==7.4.0; extra == "tests"
Requires-Dist: deepdiff; extra == "tests"
Provides-Extra: dev
Requires-Dist: lighteval[accelerate,extended_tasks,math,multilingual,quality,tests,vllm]; extra == "dev"
Provides-Extra: docs
Requires-Dist: hf-doc-builder; extra == "docs"
Requires-Dist: watchdog; extra == "docs"
Provides-Extra: extended-tasks
Requires-Dist: langdetect; extra == "extended-tasks"
Requires-Dist: openai==1.55.2; extra == "extended-tasks"
Requires-Dist: tiktoken; extra == "extended-tasks"
Provides-Extra: s3
Requires-Dist: s3fs; extra == "s3"
Provides-Extra: multilingual
Requires-Dist: stanza; extra == "multilingual"
Requires-Dist: spacy[ja,ko,th]; extra == "multilingual"
Requires-Dist: jieba; extra == "multilingual"
Requires-Dist: pyvi; extra == "multilingual"
Provides-Extra: math
Requires-Dist: latex2sympy2_extended==1.0.6; extra == "math"
Provides-Extra: wandb
Requires-Dist: wandb; extra == "wandb"
Dynamic: license-file

<p align="center">
  <br/>
    <img alt="lighteval library logo" src="./assets/lighteval-doc.svg" width="376" height="59" style="max-width: 100%;">
  <br/>
</p>


<p align="center">
    <i>Your go-to toolkit for lightning-fast, flexible LLM evaluation, from Hugging Face's Leaderboard and Evals Team.</i>
</p>

<div align="center">

[![Tests](https://github.com/huggingface/lighteval/actions/workflows/tests.yaml/badge.svg?branch=main)](https://github.com/huggingface/lighteval/actions/workflows/tests.yaml?query=branch%3Amain)
[![Quality](https://github.com/huggingface/lighteval/actions/workflows/quality.yaml/badge.svg?branch=main)](https://github.com/huggingface/lighteval/actions/workflows/quality.yaml?query=branch%3Amain)
[![Python versions](https://img.shields.io/pypi/pyversions/lighteval)](https://www.python.org/downloads/)
[![License](https://img.shields.io/badge/License-MIT-green.svg)](https://github.com/huggingface/lighteval/blob/main/LICENSE)
[![Version](https://img.shields.io/pypi/v/lighteval)](https://pypi.org/project/lighteval/)

</div>

---

**Documentation**: <a href="https://huggingface.co/docs/lighteval/main/en/index" target="_blank">HF's doc</a>

---

### Unlock the Power of LLM Evaluation with Lighteval 🚀

**Lighteval** is your all-in-one toolkit for evaluating LLMs across multiple
backends—whether it's
[transformers](https://github.com/huggingface/transformers),
[tgi](https://github.com/huggingface/text-generation-inference),
[vllm](https://github.com/vllm-project/vllm), or
[nanotron](https://github.com/huggingface/nanotron)—with
ease. Dive deep into your model’s performance by saving and exploring detailed,
sample-by-sample results to debug and see how your models stack-up.

Customization at your fingertips: letting you either browse all our existing [tasks](https://huggingface.co/docs/lighteval/available-tasks) and [metrics](https://huggingface.co/docs/lighteval/metric-list) or effortlessly create your own [custom task](https://huggingface.co/docs/lighteval/adding-a-custom-task) and [custom metric](https://huggingface.co/docs/lighteval/adding-a-new-metric), tailored to your needs.

Seamlessly experiment, benchmark, and store your results on the Hugging Face
Hub, S3, or locally.


## 🔑 Key Features

- **Speed**: [Use vllm as backend for fast evals](https://huggingface.co/docs/lighteval/use-vllm-as-backend).
- **Completeness**: [Use the accelerate backend to launch any models hosted on Hugging Face](https://huggingface.co/docs/lighteval/quicktour#accelerate).
- **Seamless Storage**: [Save results in S3 or Hugging Face Datasets](https://huggingface.co/docs/lighteval/saving-and-reading-results).
- **Python API**: [Simple integration with the Python API](https://huggingface.co/docs/lighteval/using-the-python-api).
- **Custom Tasks**: [Easily add custom tasks](https://huggingface.co/docs/lighteval/adding-a-custom-task).
- **Versatility**: Tons of [metrics](https://huggingface.co/docs/lighteval/metric-list) and [tasks](https://huggingface.co/docs/lighteval/available-tasks) ready to go.


## ⚡️ Installation

Note that lighteval is currently completely untested on Windows, and we don't support it yet. (Should be fully functional on Mac/Linux)

```bash
pip install lighteval
```

Lighteval allows for many extras when installing, see [here](https://huggingface.co/docs/lighteval/installation) for a complete list.

If you want to push results to the Hugging Face Hub, add your access token as
an environment variable:

```shell
huggingface-cli login
```

## 🚀 Quickstart

Lighteval offers the following entry points for model evaluation:

- `lighteval accelerate` : evaluate models on CPU or one or more GPUs using [🤗
  Accelerate](https://github.com/huggingface/accelerate)
- `lighteval nanotron`: evaluate models in distributed settings using [⚡️
  Nanotron](https://github.com/huggingface/nanotron)
- `lighteval vllm`: evaluate models on one or more GPUs using [🚀
  VLLM](https://github.com/vllm-project/vllm)
- `lighteval endpoint`
    - `inference-endpoint`: evaluate models on one or more GPUs using [🔗
  Inference Endpoint](https://huggingface.co/inference-endpoints/dedicated)
    - `tgi`: evaluate models on one or more GPUs using [🔗 Text Generation Inference](https://huggingface.co/docs/text-generation-inference/en/index)
    - `openai`: evaluate models on one or more GPUs using [🔗 OpenAI API](https://platform.openai.com/)

Here’s a quick command to evaluate using the Accelerate backend:

```shell
lighteval accelerate \
    "model_name=gpt2" \
    "leaderboard|truthfulqa:mc|0|0"
```

## 🙏 Acknowledgements

Lighteval started as an extension of the fantastic [Eleuther AI
Harness](https://github.com/EleutherAI/lm-evaluation-harness) (which powers the
[Open LLM
Leaderboard](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard))
and draws inspiration from the amazing
[HELM](https://crfm.stanford.edu/helm/latest/) framework.

While evolving Lighteval into its own standalone tool, we are grateful to the
Harness and HELM teams for their pioneering work on LLM evaluations.

## 🌟 Contributions Welcome 💙💚💛💜🧡

Got ideas? Found a bug? Want to add a
[task](https://huggingface.co/docs/lighteval/adding-a-custom-task) or
[metric](https://huggingface.co/docs/lighteval/adding-a-new-metric)?
Contributions are warmly welcomed!

If you're adding a new feature, please open an issue first.

If you open a PR, don't forget to run the styling!

```bash
pip install -e .[dev]
pre-commit install
pre-commit run --all-files
```
## 📜 Citation

```bibtex
@misc{lighteval,
  author = {Habib, Nathan and Fourrier, Clémentine and Kydlíček, Hynek and Wolf, Thomas and Tunstall, Lewis},
  title = {LightEval: A lightweight framework for LLM evaluation},
  year = {2023},
  version = {0.8.0},
  url = {https://github.com/huggingface/lighteval}
}
```
