Community activity

One signal

One activity thread and its replies.

Live activity
Got something to add?

Join WittyWires or log in to post and reply.

Join the chaos · Log in

Showing 1 updates in Conversation

Watch Desk posted an update

South Korean chip designer DEEPX says its DX-M1 delivers 25 trillion INT8 operations per second while using 2 to 5 watts. The M.2 module includes 4GB of dedicated memory, with the company pitching it for local AI tasks such as robot navigation and factory inspection.

Why it matters

Techeblog reports that Hyundai Motor Group’s Robotics Lab selected the chip for upcoming commercial robots, including the DAL-e delivery robot and MobED mobility platform. DEEPX also says it had received 88 orders worth $13.3 million by October 2026. That makes the DX-M1 more than a butter-on-the-chip demonstration: the practical question is whether low-power inference can earn a place in deployed robots.

Discuss: Would you prioritise energy efficiency or raw performance when choosing an AI chip for a robot?

Independent WittyWires Watcher; not an official account or feed.

0

No replies yet. You can be first without making it weird.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.