Been out all day so I missed a ton of @axisrobotics tasks 😭 Finally got the chance to grind at around 8PM SGT, so I borrowed my nephew’s iPad just so I could get some tasks in ✨ First time doing Axis on an iPad too and damn… it was HARD for me. Only managed to do a few tasks 😭 At least I had a live singer in the background while struggling 🦾🎶 We vibin while we grind. How many tasks did you guys get done today? 👀 gAxis fam!

Thrilled our paper has been accepted to the Physical World Models (PWMS) workshop at #IROS2026# 🎉 Growable, community-driven data engine for robot manipulation: 1,800+ tasks, 1.5M+ trajs. π0.5+AXIS: 83.9→88.8 on LIBERO-Plus. See you in Pittsburgh Oct 1 👋 Links ⬇️

Most people hear @axisrobotics and think “another robotics project” but Axis is actually working on something much deeper: the data infrastructure that helps robots learn from the real world. I spent some time digging into how it works and why it could matter in everyday life. here’s the simple version 🧵👇

A community report flagged an Axis user who created 14 sub-addresses, all detected by our algorithm and granted the Axis Challenger badge. Full verification confirmed all data was manually completed by the user — no scripts or bot manipulation, and the contributions are valid for policy training. After review, we will not ban this user: ✅ Data is genuine & usable ✅ No automated cheating ✅ Multi-address participation is allowed; multiple addresses alone do not negatively impact the project. 经社区举报，发现某axis用户用一个地址创建了14个子地址，并且这14个地址都被算法识别，给予了axis challenger badge。 经过仔细核验，14个地址的数据都由该用户手动完成，不存在脚本和bot manipulation痕迹。贡献的数据也是可用于policy training的。 经过讨论，我们决定不封杀该用户。 原因1）贡献的数据真实可用。2）没有使用脚本作弊。3）多个地址参与做任务的行为是被允许的，多地址本身并不影响项目发展。

huge Coinbase is paying more and attention to APAC regions and this a very positive signal for APAC builders to start building on Base. I will write a tutorial on how to build on base soon. But Apac builder can start by following @baseapac account

The interesting part about @axisrobotics is not just having more data. Same number of trajectories, but Dataset V1 pushed LIBERO Plus from 57.5 to 88.8. The difference came from better coverage, cleaning, and randomization across camera, noise, and layout. 207 tasks, 50K plus human demos, 60K plus scene variants, and everything is open. That is a much stronger signal than simply saying they collected more data.

BASECAST/002: ACCELERATING THE NEXT GENERATION OF ROBOTICS ft. @chris_anm0, Founder & Core contributor @axisrobotics Physical AI won’t be built on models alone, it will be built on data from the real world. Key Timestamps: 1:03 - Chris’ builder journey before Axis 4:09 - What is Axis Robotics? 7:46 - The user journey on Axis Hub 11:06 - The train-to-earn model 12:25 - Competition and the hybrid data advantage 17:14 - The future of Physical AI 19:10 - Industry collaboration and the Axis roadmap 21:00 - Advice for the builders of tomorrow

The on question on every newbie and web3 folks right now 👇 How do I make the most from the robotic meta and what project is popping on that narrative 👀 I have an answer: put the avocado beside the mango Generational cook for 2026 is @axisrobotics Get your roles, finish all task and thank me later. Get in: Everything will make sense soon, see you on the side that prints

Over 2000 bookmarks on my previous video on how to Play Axis Robotic 😳😧 So I decided to bless you all with a new 2 minutes interaction with @axisrobotics and how easy some training can be… As of today I think I’m the best Robotic trainer, I should be hired by Elon Musk.

Watching Axis community training those Franka arms every day makes me think that in the future, more and more repetitive work in the physical world will be replaced by machines, and what is left for human beings to think about is: love, longevity, and all sorts of creative work.