# anuj0456/OpenArch

> Educational PyTorch implementations of modern open-source LLM architectures (Llama, Qwen, DeepSeek, Gemma, GPT-OSS, Kimi, and more), written from scratch with one readable file per model for comparing architectural choices.

- Magnitude: 0.4 out of 10 — Quiet
- Stars: 421 total · +0 stars on 2026-10-01 (UTC), ≈ 0 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: Python · License: MIT · Created: 2026-05-27 · Last push: 2026-09-30
- GitHub: https://github.com/anuj0456/OpenArch · Page: https://gitnova.dev/en/r/anuj0456/OpenArch

## Useful for

- Study how attention, MoE, and normalization work in a specific model from a single file
- Compare architectural choices across LLMs side by side for an article or talk
- Contribute an implementation of an uncovered model from the LLM Architecture Gallery

## Why it’s here

- Hacker News: “OpenArch – PyTorch implementations of modern LLM architectures” — 139 points, 17 days ago.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 34
- Issues and pull requests: 24
- Watchers: 3
- Average over the last week: 1 per day
- Usual pace: 17 per day

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-10-04 … 2026-11-02: None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None

## Magnitude by day, last 14 days

09-18 2.9, 09-19 2.5, 09-20 1.8, 09-21 1.5, 09-22 1.5, 09-23 1.5, 09-24 1.3, 09-25 1.3, 09-26 1.3, 09-27 0.7, 09-28 1.1, 09-29 1.1, 09-30 1.0, 10-01 0.5

## Hacker News

- OpenArch – PyTorch implementations of modern LLM architectures — 139 points, 32 comments: https://news.ycombinator.com/item?id=49693384

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-01 08:02 UTC, updated every 30 minutes.