Dev Radar
Support
LiveUpdated 2026-10-30 21:12 UTC

minimind

minimind — 🧠 Train a 64M-parameter LLM from scratch in just 2h! - jingyaogong/minimind

minimind is 🧠 Train a 64M-parameter LLM from scratch in just 2h! - jingyaogong/minimind. It is ranked #2043 on the Dev Radar, in AI dev tools, first seen 17 days ago and shared in 2 posts (3.6k views).

Visit github.com

GitHub - jingyaogong/minimind: 🧠 Train a 64M-parameter LLM from scratch in just 2h! · GitHub Skip to content Navigation Menu Sign in Appearance settings Search / Sign in Sign up Appearance settings You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert {{ message }} Repository navigation…

What people said about minimind on X

想搞懂大模型怎么训练?这个开源项目把全流程摊开了。 MiniMind 用原生 PyTorch 实现了从预训练到后训练的多种流程,不只教你调用模型,还能直接研究训练代码。 🧠 预训练、SFT、LoRA、DPO ⚙️ PPO / GRPO、工具调用与 Agent 强化学习 💻 代码相对精简,适合边读边跑、动手实验 项目介绍中提到,64M 参数模型可用较低成本完成训练。不过实际耗时和费用取决于硬件、数据规模及训练配置,不能直接照搬宣传数字。 如果你想从代码层面理解大模型如何训练,这个仓库值得收藏。 🔗 https://github.com/jingyaogong/minimind

— @bkdgiffug, 7 days ago · 39 likes · see the post

花 4 元、用 2 小时,就能从零训练一个 64M 小模型? 开源项目 MiniMind 用原生 PyTorch 实现了预训练、SFT、LoRA、DPO、PPO/GRPO 及 Agent 训练等流程,模型结构对齐 Qwen3,个人也能动手复现。 想搞懂大模型训练原理,这个项目值得研究。 🔗 GitHub:https://github.com/jingyaogong/minimind

— @bkdgiffug, 17 days ago · 15 likes · see the post

Alternatives to minimind

minimind in numbers

FAQ

What is minimind?

🧠 Train a 64M-parameter LLM from scratch in just 2h! - jingyaogong/minimind It was first shared on X 17 days ago and is ranked #2043 on the Dev Radar.

Is minimind free?

It is open source.

Who shared minimind?

1 account on X, including @bkdgiffug, in 2 posts totalling 3.6k views.

Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.2k posts from 5k X accounts over the last 21 days, 8.8k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-10-30 21:12 UTC. Full method.