Source: https://seerepo.com/compare/harvard-edge/cs249r_book/vs/OptimalScale/LMFlow
Markdown source: https://seerepo.com/compare/harvard-edge/cs249r_book/vs/OptimalScale/LMFlow.md
Description: cs249r_book and LMFlow share 18 technologies: pyzmq, pyarrow, jupyter-core, tornado, cycler, fonttools, contourpy, matplotlib-inline, prompt-toolkit, matplotlib, torch, transformers, torchvision, jinja2, markupsafe, pillow, Pydantic, and psutil. Both are written in Python.

# cs249r_book vs LMFlow

cs249r_book and LMFlow share 18 technologies: pyzmq, pyarrow, jupyter-core, tornado, cycler, fonttools, contourpy, matplotlib-inline, prompt-toolkit, matplotlib, torch, transformers, torchvision, jinja2, markupsafe, pillow, Pydantic, and psutil. Both are written in Python.

- [harvard-edge/cs249r_book](https://seerepo.com/r/harvard-edge/cs249r_book): Machine Learning Systems: Foundations, Scaling, Agentic AI, and Physical AI (Vols I–IV) • Harvard CS249r | https://mlsysbook.ai
- [OptimalScale/LMFlow](https://seerepo.com/r/OptimalScale/LMFlow): An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Models for All.

## What they share and what sets them apart

- In common: pyzmq, pyarrow, jupyter-core, tornado, cycler, fonttools, contourpy, matplotlib-inline, prompt-toolkit, matplotlib, torch, transformers, torchvision, jinja2, markupsafe, pillow, Pydantic, psutil
- Only in cs249r_book: async-lru, flexcache, requests-cache, anyio, jupyterlab-server, marimo, Vite, React, idb, Next.js, jupyterlab-pygments, soupsieve, tinycss2, ipywidgets, jupyter-server-terminals, jupyterlab-widgets, seaborn, widgetsnbextension, Rich, Plotly, argon2-cffi, argon2-cffi-bindings, groq, Ollama, OpenAI, tflite, scikit-learn, sentence-transformers, librosa, nltk, ogb, torch-geometric, betterbib, bleach, defusedxml, ghostscript, latexcodec, markdown-it-py, mdurl, mistune, nbconvert, pandocfilters, pybtex, pylatexenc, pypandoc, python-frontmatter, sgf, tablib, webencodings, xlrd, xlwt, xmltodict, beautifulsoup4, lxml, epubcheck, bibtexparser, jupytext, nbformat, annotated-types, cattrs, fastjsonschema, fqdn, jsonschema-specifications, linkml, referencing, rfc3339-validator, rfc3986-validator, rfc3987-syntax, jsonschema, linkml-runtime, prometheus-client, python-json-logger, Tailwind CSS, Chart.js, D3, Mermaid, ink-mde, compromise, dompurify, katex, markdown-it, markdown-it-container, jspdf, jspdf-autotable, Framer, react-medium-image-zoom, Lucide
- Only in LMFlow: pickleshare, Hugging Face, cpm-kernels, einops, NVIDIA, open-clip-torch, sentencepiece, timm, tokenizers, triton, datasets, accelerate, diffusers, peft, bitsandbytes, xformers, evaluate, tensordict

## Side by side

| | cs249r_book | LMFlow |
|---|---|---|
| Size | 828,618 lines written, 1,983 code files | 35,146 lines written, 108 code files |
| Who writes it | 2.7% of its commits signed by AI. With Codex, Claude Code, and Cursor | 0% of its commits signed by AI |
| Activity | Last push November 15, 2026, 22,406 commits, 30,209 stars | Last push August 10, 2026, 1,262 commits, 8,481 stars |
| Tests | 632 test files. 16.8% of its code files have a test that imports them | 22 test files. 16.7% of its code files have a test that imports them |

- 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, cs249r_book or LMFlow?

cs249r_book, with 828,618 lines written; LMFlow has 35,146.

The proof: [harvard-edge/cs249r_book's code at commit 5b6b3c8](https://github.com/harvard-edge/cs249r_book/tree/5b6b3c8d82e429ea25c9c04dd1be477937008c28)

### Which has more GitHub stars, cs249r_book or LMFlow?

cs249r_book, with 30,209 stars; LMFlow has 8,481.

The proof: [harvard-edge/cs249r_book on GitHub](https://github.com/harvard-edge/cs249r_book)

### Which has more tests, cs249r_book or LMFlow?

cs249r_book, with 632 test files; LMFlow has 22.

The proof: [harvard-edge/cs249r_book's code at commit 5b6b3c8](https://github.com/harvard-edge/cs249r_book/tree/5b6b3c8d82e429ea25c9c04dd1be477937008c28)

### Which was updated more recently, cs249r_book or LMFlow?

cs249r_book: its last push was on November 15, 2026, and LMFlow's on August 10, 2026.

The proof: [harvard-edge/cs249r_book's commits on GitHub](https://github.com/harvard-edge/cs249r_book/commits)

### Which has more commits signed by AI, cs249r_book or LMFlow?

cs249r_book: an AI signed 2.7% of its commits, with Codex, Claude Code, and Cursor.

The proof: [the commits each tool signed in harvard-edge/cs249r_book](https://seerepo.com/r/harvard-edge/cs249r_book/ai)

## Sources

- harvard-edge/cs249r_book: Read on November 15, 2026, at commit 5b6b3c8 (https://github.com/harvard-edge/cs249r_book/tree/5b6b3c8d82e429ea25c9c04dd1be477937008c28)
- OptimalScale/LMFlow: Read on November 17, 2026, at commit e9b1da0 (https://github.com/OptimalScale/LMFlow/tree/e9b1da012f08d17fadad698278f505800d30a8af)

## 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.
