mini-agi
mini-agi is Continual learning model trained from scratch on 8GB VRAM laptop with batch-1 stream of data. - volotat/mini-AGI. It is ranked #396 on the AI Radar, in AI research, first seen 1 h ago and shared in 1 post (596 views).
GitHub - volotat/mini-AGI: Continual learning model trained from scratch on 8GB VRAM laptop with batch-1 stream of data. · 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 }} volotat / mini-AGI Public Notifications You must be signed in to change notification settings Fork 22 Star 316 main Branches Tags Go to file Code Open more…
What people said about mini-agi on X
🧫 离谱，有人在一张 8GB 显存的笔记本显卡上，从零训出了一个会「边读边学」的模型 mini-AGI。 仓库 9 月 19 号才建起来，作者是 Alexey Borsky，代码主要由 Claude Opus 5 写完、调完、验完。 现在的模型基本都是训完就定型：预训练烧完一轮算力，权重冻住，之后你跟它说什么它都不会真的记住，全靠把上下文一次次喂回去。mini-AGI 反着来，它一直在读，读到的东西直接进参数，读完 3.18 亿个字符之后，内部的专家池已经自己长到了 169 个。 最卡人的显存问题它是这么绕开的：权重就是磁盘上的普通文件，用到哪块再分页调进显卡，所以模型能长多大取决于你硬盘还剩多少，而不是显卡有几个…
— @Ryrenz, 1 h ago · 8 likes · see the post
Alternatives to mini-agi
- arxiv-complete — We’re on a journey to advance and democratize artificial intelligence through open source and open science.
- klpo — KL-Regularized Policy Optimization for Agentic Reinforcement Learning - yifanzhang-pro/KLPO
- bindcraft2 — Easy to use suite for designing a wide range of protein binder modalities. - PacesaLab/BindCraft2
- recurrent-looped-tranformer — Official Project Page for Recurrent Looped Transformer (RLT) - yifanzhang-pro/recurrent-looped-tranformer
- VLA-Replica — Web home for VLAReplica Benchmark for Robotics
- humanbrain-pollard-h01 — We’re on a journey to advance and democratize artificial intelligence through open source and open science.
mini-agi in numbers
- Rank on the AI Radar: #396 of 1703
- Shared in 1 post by 1 account: @Ryrenz
- 596 views on those posts
- First seen 1 h ago, last shared 1 h ago
- Pricing seen by Jev: open source
- Market: AI research
FAQ
What is mini-agi?
Continual learning model trained from scratch on 8GB VRAM laptop with batch-1 stream of data. - volotat/mini-AGI It was first shared on X 1 h ago and is ranked #396 on the AI Radar.
Is mini-agi free?
It is open source.
Who shared mini-agi?
1 account on X, including @Ryrenz, in 1 post totalling 596 views.
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 41k posts from 5k X accounts over the last 14 days, 1.7k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-22 04:47 UTC. Full method.