Source: https://seerepo.com/r/EleutherAI/lm-evaluation-harness/built
Markdown source: https://seerepo.com/r/EleutherAI/lm-evaluation-harness/built.md
Description: lm-evaluation-harness is a C++ repo. It uses AI through evaluate, fugashi, rouge-score, scikit-learn, datasets, and 1 more and makes its content with jinja2.

# What is lm-evaluation-harness built with?

lm-evaluation-harness is a C++ repo. It uses AI through evaluate, fugashi, rouge-score, scikit-learn, datasets, and 1 more and makes its content with jinja2. It has 107,687 written lines, with 51 test files.

Read on November 9, 2026, at commit d6de816.

Repository: https://github.com/EleutherAI/lm-evaluation-harness

## Backend

### AI

- **[evaluate](https://seerepo.com/uses/evaluate)**
- **fugashi**
- **[rouge-score](https://seerepo.com/uses/rouge-score)**
- **[scikit-learn](https://seerepo.com/uses/scikit-learn)**
- **[datasets](https://seerepo.com/uses/datasets)**: 1 file
- **sacrebleu**: 1 file

## Support

### Content

- **[jinja2](https://seerepo.com/uses/jinja2)**

### Tools

- **typing-extensions**

### Other

- **dill**
- **emoji**
- **more-itertools**
- **tqdm**
- **word2number**
- **NumPy**: 8 files

## Other repos with a similar stack

- [ray-project/ray](https://seerepo.com/r/ray-project/ray/built): rouge-score, scikit-learn, datasets, sacrebleu, jinja2 · [Compare](https://seerepo.com/compare/EleutherAI/lm-evaluation-harness/vs/ray-project/ray)
- [mlflow/mlflow](https://seerepo.com/r/mlflow/mlflow/built): evaluate, rouge-score, scikit-learn, datasets, jinja2 · [Compare](https://seerepo.com/compare/EleutherAI/lm-evaluation-harness/vs/mlflow/mlflow)
- [NVIDIA/TensorRT-LLM](https://seerepo.com/r/NVIDIA/TensorRT-LLM/built): evaluate, rouge-score, datasets, jinja2 · [Compare](https://seerepo.com/compare/EleutherAI/lm-evaluation-harness/vs/NVIDIA/TensorRT-LLM)
- [datawhalechina/hello-agents](https://seerepo.com/r/datawhalechina/hello-agents/built): scikit-learn, datasets, jinja2 · [Compare](https://seerepo.com/compare/datawhalechina/hello-agents/vs/EleutherAI/lm-evaluation-harness)
- [unslothai/unsloth](https://seerepo.com/r/unslothai/unsloth/built): scikit-learn, datasets, jinja2 · [Compare](https://seerepo.com/compare/EleutherAI/lm-evaluation-harness/vs/unslothai/unsloth)
- [jingyaogong/minimind](https://seerepo.com/r/jingyaogong/minimind/built): scikit-learn, datasets, jinja2 · [Compare](https://seerepo.com/compare/EleutherAI/lm-evaluation-harness/vs/jingyaogong/minimind)

## Frequently asked questions

### What language is lm-evaluation-harness written in?

lm-evaluation-harness is written in C++, with 107,687 written lines.

### Does lm-evaluation-harness use AI?

Yes. lm-evaluation-harness uses AI through evaluate, fugashi, rouge-score, scikit-learn, datasets, and sacrebleu.

### Does lm-evaluation-harness have tests?

Yes. lm-evaluation-harness has 51 test files.

### Which repos have a stack like lm-evaluation-harness's?

The ones that share most of its stack: ray-project/ray, mlflow/mlflow, and NVIDIA/TensorRT-LLM.

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.
