Source: https://seerepo.com/compare/EleutherAI/lm-evaluation-harness/vs/jingyaogong/minimind
Markdown source: https://seerepo.com/compare/EleutherAI/lm-evaluation-harness/vs/jingyaogong/minimind.md
Description: lm-evaluation-harness and minimind share 3 technologies: scikit-learn, datasets, and jinja2. lm-evaluation-harness is written in C++ and minimind in Python.

# lm-evaluation-harness vs minimind

lm-evaluation-harness and minimind share 3 technologies: scikit-learn, datasets, and jinja2. lm-evaluation-harness is written in C++ and minimind in Python.

- [EleutherAI/lm-evaluation-harness](https://seerepo.com/r/EleutherAI/lm-evaluation-harness): A framework for few-shot evaluation of language models.
- [jingyaogong/minimind](https://seerepo.com/r/jingyaogong/minimind): 🧠 Train a 64M-parameter LLM from scratch in just 2h!

## What they share and what sets them apart

- In common: scikit-learn, datasets, jinja2
- Only in lm-evaluation-harness: evaluate, fugashi, rouge-score, sacrebleu
- Only in minimind: FastAPI, Flask, flask-cors, Streamlit, uvicorn, Rich, einops, ModelScope, nltk, OpenAI, sentence-transformers, sentencepiece, tiktoken, trl, transformers, jsonlines, marshmallow, Pydantic, ngrok, psutil, swanlab

## Side by side

| | lm-evaluation-harness | minimind |
|---|---|---|
| Size | 107,687 lines written, 2 code files | 4,696 lines written, 15 code files |
| Who writes it | 0.3% of its commits signed by AI. With Claude Code and Cursor | 0.8% of its commits signed by AI. With Codex, Claude Code, and Cursor |
| Activity | Last push November 4, 2026, 3,347 commits, 14,426 stars | Last push October 27, 2026, 382 commits, 65,173 stars |
| Tests | 51 test files | 0 test files |

- Size: Lines written by people and agents, without generated ones; code files, without tests.
- Who writes it: A commit counts as AI's when it carries its signature: a Co-Authored-By line, its author name or the merge of its branch.
- Activity: From GitHub and its history, on the day SeeRepo read it.
- Tests: The files that test the code, and the share of code files a test imports.

## Frequently asked questions

### Which is bigger, lm-evaluation-harness or minimind?

lm-evaluation-harness, with 107,687 lines written; minimind has 4,696.

The proof: [EleutherAI/lm-evaluation-harness's code at commit d6de816](https://github.com/EleutherAI/lm-evaluation-harness/tree/d6de81643928d653435c431bae19945d41d32520)

### Which has more GitHub stars, lm-evaluation-harness or minimind?

minimind, with 65,173 stars; lm-evaluation-harness has 14,426.

The proof: [jingyaogong/minimind on GitHub](https://github.com/jingyaogong/minimind)

### Which has more tests, lm-evaluation-harness or minimind?

lm-evaluation-harness, with 51 test files.

The proof: [EleutherAI/lm-evaluation-harness's code at commit d6de816](https://github.com/EleutherAI/lm-evaluation-harness/tree/d6de81643928d653435c431bae19945d41d32520)

### Which was updated more recently, lm-evaluation-harness or minimind?

lm-evaluation-harness: its last push was on November 4, 2026, and minimind's on October 27, 2026.

The proof: [EleutherAI/lm-evaluation-harness's commits on GitHub](https://github.com/EleutherAI/lm-evaluation-harness/commits)

### Which has more commits signed by AI, lm-evaluation-harness or minimind?

minimind: an AI signed 0.8% of its commits, with Codex, Claude Code, and Cursor; in lm-evaluation-harness, 0.3%.

## Sources

- EleutherAI/lm-evaluation-harness: Read on November 9, 2026, at commit d6de816 (https://github.com/EleutherAI/lm-evaluation-harness/tree/d6de81643928d653435c431bae19945d41d32520)
- jingyaogong/minimind: Read on November 12, 2026, at commit 233d00f (https://github.com/jingyaogong/minimind/tree/233d00fa068ec2b8c0d4f07b703934d6391a5a73)

## 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.
