jingyaogong/minimind
🧠 Train a 64M-parameter LLM from scratch in just 2h!
About the project
An open-source educational project for training a tiny 64M-parameter LLM (MiniMind) from scratch in PyTorch, covering the full pipeline: pretrain, SFT, LoRA, RLHF, RLAIF and Tool Use.
Useful for
- Train your own tiny LLM from scratch on a single GPU in a couple of hours
- Study pretrain, SFT, LoRA and RLHF implementations without transformers/trl wrappers
- Deploy an OpenAI-API-compatible MiniMind server for chat UIs
README summarized by DeepSeek V4.1 Flash. Details may be inaccurate.
Why it’s trending
- 82 stars so far today, about 100 expected by the end of the day.
- Interest is fading: the two-day pace is 47% of the previous week and a half.
- GitHub Trending this month: #15, +6,867 stars.
- About 16 forks a day — people are taking the code.
Stars per day
Bars are daily stars, the line is the usual pace. Red marks spike days.
Numbers
- Total stars
- 61,588
- Stars in a day
- 100
- Forks
- 8,011
- Issues and pull requests
- 775
- Watchers
- 278
- Language
- Python
- License
- Apache-2.0
- Latest release
- v2 · October 21, 2025
- Created
- July 27, 2024
- Last push
- September 18, 2026
Star trust
Growth looks organic: forks and discussion are in line with active projects, and stars arrive unevenly, the way people give them.
These are heuristics, not a verdict: we judge by the repository’s behavior, not by a list of stargazers.
Spotted in
- September 18, 2026GitHub Trending this month: #15, +6,867 stars
- September 17, 2026GitHub Trending this month: #16, +6,769 stars
- September 16, 2026GitHub Trending this month: #21, +6,660 stars
Share
README badge
Paste it into your README — the badge shows the current magnitude and links to this page.
[](https://gitnova.dev/en/r/jingyaogong/minimind)Similar projects
-
8.6
TheoLeeCJ/SemIf
An open reproduction of the Jev-style semantic decision interface: reads typed option probabilities directly from a 4B model's logits without generating text. Runs on a single RTX 3090.
-
7.1
vinnylarouge/jevlike
A small trainable model that picks one option from a changing list of text options in a single forward pass, an open alternative to TypeSafe's Jev. Aimed at developers who need fast classification or menu selection…
-
6.9
Continuum-AI-Corp/OrcaBonsai-27B-Uncensored
A tool for runtime refusal ablation in the compressed Ternary Bonsai 2 27B LLM: it projects the residual stream to remove the refusal direction without modifying or re-quantizing weights. Runs on Apple Silicon via MLX.
-
6.4
TianyuCodings/NanoJev
A nano replica of Jev built on Qwen3-0.6B that outputs probability distributions over candidates in a single forward pass with no token decoding, plus a training pipeline and game demos (maze, Snake).
-
6.3
zhengkid/Dream-RSI
Research project on recursive self-improvement: past discovery histories become a replay simulator used to evaluate and refine exploration policies. Code is still being prepared; paper and demo are available.
-
5.0
ekzhang/openjev-sglang
HTTP server implementing the TypeSafe/Jev API for structured text classification on Qwen3.6-35B-A3B via SGLang; returns answer probabilities without generating a chain of thought.