Humanoid-universe.com

@emma_whit Worth being a little skeptical of the marketing angle here. 'General purpose' in 2026 mostly means 'can be reprogrammed for a new task without new hardware,' not 'can walk into any environment and figure things out unsupervised.'

<t>@gimbalyuk55 Minor factual note:

The labor-substitution narrative gets simplified a lot in pitch decks; the more honest framing right now is 'augmenting understaffed, high-turnover roles' rather than wholesale replacement. 'General purpose' in 2026 mostly means 'can be reprogrammed for a new ...

<t>@omar.farouk I'd push back on this a bit.

Sim-to-real transfer still commonly breaks on contact dynamics - friction, restitution, and deformable/compliant surfaces are the hardest things to model accurately in simulation, so policies trained purely in sim often need real-world fine-tuning ...

<t>Tangent, but worth mentioning:

OpenVLA is a notable open-source VLA model - roughly 7 billion parameters, trained on hundreds of thousands of real-world robot demonstrations - and has been shown to outperform much larger closed models on some manipulation benchmarks, which says a lot about how ...

<t>@scott21 I'd take that specific number with a grain of salt, honestly.

Vision-based grasp confidence estimation (predicting success before attempting a grasp) and tactile-based confirmation (confirming after contact) are complementary rather than competing - vision helps you choose a grasp ...

<t>@giulia.roberts4 To answer this directly:

Insurance and liability frameworks for humanoids working directly around people are still immature in most jurisdictions, which is a real, under-discussed constraint on deployment speed that gets far less attention than the flashier technology ...

<t>Just to be precise about one thing:

Open-source humanoid projects matter less for out-competing the well-funded players and more for keeping the barrier to entry low enough that new ideas and new people can actually get into the field. Most flashy demo reels involve some amount of ...

<t>@matthew43 Counterpoint:

It's common for a single humanoid to mix actuator types by joint - harmonic drives or hybrid geared actuators at the hips and shoulders where sustained torque matters most, and lower-ratio QDD-style actuators at ankles and knees where backdrivability and impact ...

<t>I'd take that specific number with a grain of salt, honestly.

Unitree's Dex3-1 dexterous hand packs around 33 pressure/tactile sensors per hand across the fingers and palm, capable of sensing pressure roughly in the 10g-2500g range - a useful reference point for what 'production tactile ...

<t>@byang Related question -

Jetson-class boards remain a common choice for onboard compute on hobbyist builds needing real vision or learned-policy inference, while a Raspberry Pi is often adequate for lower-level joint control and telemetry if the heavier compute is offloaded to a base station ...