Source: https://seerepo.com/compare/FareedKhan-dev/train-llm-from-scratch/vs/hpcaitech/ColossalAI
Markdown source: https://seerepo.com/compare/FareedKhan-dev/train-llm-from-scratch/vs/hpcaitech/ColossalAI.md
Description: train-llm-from-scratch and ColossalAI share 5 technologies: torchaudio, torchvision, torch, tiktoken, and datasets. Both are written in Python.

# train-llm-from-scratch vs ColossalAI

train-llm-from-scratch and ColossalAI share 5 technologies: torchaudio, torchvision, torch, tiktoken, and datasets. Both are written in Python.

- [FareedKhan-dev/train-llm-from-scratch](https://seerepo.com/r/FareedKhan-dev/train-llm-from-scratch): A straightforward method for training your LLM, from downloading data to generating text.
- [hpcaitech/ColossalAI](https://seerepo.com/r/hpcaitech/ColossalAI): Making large AI models cheaper, faster and more accessible

## What they share and what sets them apart

- In common: torchaudio, torchvision, torch, tiktoken, datasets
- Only in train-llm-from-scratch: h5py
- Only in ColossalAI: FastAPI, uvicorn, chromadb, Hugging Face, anyio, Ray, Gradio, seaborn, matplotlib, Rich, SQLAlchemy, ModelScope, OpenAI, scikit-learn, sentence-transformers, transformers, LangChain, triton, peft, diffusers, flash-attn, timm, safetensors, bitsandbytes, vLLM, einops, accelerate, galore-torch, math-verify, sentencepiece, tokenizers, jinja2, markdown-it-py, markupsafe, mdurl, pypdf, unstructured, pillow, jsonlines, annotated-types, pydantic-core, Pydantic, psutil

## Side by side

| | train-llm-from-scratch | ColossalAI |
|---|---|---|
| Size | 12,346 lines written, 79 code files | 317,403 lines written, 974 code files |
| Who writes it | 7.1% of its commits signed by AI. With Claude Code and GitHub Copilot | 0.3% of its commits signed by AI. With Codex and Claude Code |
| Activity | Last push October 2, 2026, 155 commits, 13,179 stars | Last push November 6, 2026, 3,887 commits, 41,430 stars |
| Tests | 26 test files. 39.2% of its code files have a test that imports them | 505 test files. 1.1% 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, train-llm-from-scratch or ColossalAI?

ColossalAI, with 317,403 lines written; train-llm-from-scratch has 12,346.

The proof: [hpcaitech/ColossalAI's code at commit da8f5c2](https://github.com/hpcaitech/ColossalAI/tree/da8f5c2dd56f97d3d23d3d222b114e1a3d4710f5)

### Which has more GitHub stars, train-llm-from-scratch or ColossalAI?

ColossalAI, with 41,430 stars; train-llm-from-scratch has 13,179.

The proof: [hpcaitech/ColossalAI on GitHub](https://github.com/hpcaitech/ColossalAI)

### Which has more tests, train-llm-from-scratch or ColossalAI?

ColossalAI, with 505 test files; train-llm-from-scratch has 26.

The proof: [hpcaitech/ColossalAI's code at commit da8f5c2](https://github.com/hpcaitech/ColossalAI/tree/da8f5c2dd56f97d3d23d3d222b114e1a3d4710f5)

### Which was updated more recently, train-llm-from-scratch or ColossalAI?

ColossalAI: its last push was on November 6, 2026, and train-llm-from-scratch's on October 2, 2026.

The proof: [hpcaitech/ColossalAI's commits on GitHub](https://github.com/hpcaitech/ColossalAI/commits)

### Which has more commits signed by AI, train-llm-from-scratch or ColossalAI?

train-llm-from-scratch: an AI signed 7.1% of its commits, with Claude Code and GitHub Copilot; in ColossalAI, 0.3%.

## Sources

- FareedKhan-dev/train-llm-from-scratch: Read on November 10, 2026, at commit bdd480a (https://github.com/FareedKhan-dev/train-llm-from-scratch/tree/bdd480a518094ca834e05ab32131dc126b522d3d)
- hpcaitech/ColossalAI: Read on November 6, 2026, at commit da8f5c2 (https://github.com/hpcaitech/ColossalAI/tree/da8f5c2dd56f97d3d23d3d222b114e1a3d4710f5)

## 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.