For robotics teams that need to ship fast

Add any skill to any robot — in under 60 minutes

AI pipeline. Any robot platform. No rebuilding from scratch.

See how it works →

No credit card required · Any robot platform · Powered by our research

Sound familiar?

The moment you realise you can't afford to rebuild from scratch again.

TIME PRESSURE

Your demo is in 3 days.

The robot still can't do the new task. You're staring at weeks of training work you don't have time for.

NEW REQUIREMENT

You just won the contract.

It requires a capability your robot doesn't have. The clock is ticking and building from scratch isn't an option.

FALLING BEHIND

Your competitor shipped first.

They deployed a new skill in days. You're still setting up your simulation environment.

These aren't edge cases. Every robot team hits them. The difference is how fast you recover.

See how we solve it →

The problem

Every new skill means rebuilding from zero

Here's what gets in the way.

Behaviors that don't fit your environment

Generic behaviors fall short in real environments.

Every environment is different

You rebuild from scratch instead of reusing what works.

Slow, expensive, fragmented

Training stays slow and expensive across teams.

The promise

<60min

From environment scan to deployed robot skill

39tasks

Out-of-the-box robot skills

3simulators

Supported training environments

Powered by cap-x — open-source robotics AI framework, peer-reviewed and published

View research on GitHub →

How it works

From idea to deployed skill in under 60 minutes

Three steps. No ML PhD required. Just scan, train, ship.

01~3 min

Scan your environment

Use our LiDAR scanner app to capture your real workspace. Point, scan, done — you get a simulation-ready 3D scene that matches your actual environment.

iPhone LiDAR app · 3D mapping · Instant sim export

02~40 min

AI trains the policy

Our AI agent writes and iterates Python control code in cloud simulation. No ML expertise needed — the system auto-synthesizes skills and runs reinforcement learning on your captured environment.

Code-as-Policy agents · Cloud GPU · RL fine-tuning

03~5 min

Deploy to real hardware

One click pushes the trained policy to your robot. Works across humanoids, quadrupeds, and manipulators. The skill runs on day one — no tweaking, no babysitting.

Any robot platform · One-click deploy · Cross-embodiment

Total: under 60 minutes

Works with any robot

One platform. Every form factor.

Unitree

Humanoid / Quadruped

Franka

Manipulator

Universal Robots

Manipulator

Cross-embodiment transfer — train once, deploy to multiple robot types · More platforms in active development

Why OpenKinematics

Stop rebuilding from scratch

Every robot team faces the same wall. We built the ladder over it.

Feature
OpenKinematics
Build yourself
New skill deployment
< 60 minutes
Weeks to months
Robot compatibility
Any platform
One at a time
ML expertise required
None
Senior ML engineer
Environment capture
LiDAR scan, 3 min
Manual CAD modeling
Simulation training
Cloud, automated
Custom setup, manual
Research transparency
Published research
Black box
Cost
Subscription
$200k+ eng time / skill

Research-backed

Enterprise performance.
Built on open research.

OpenKinematics is powered by cap-x — an open-source robotics AI framework developed in our labs. We publish our research so the field advances. Our platform packages it into a production-ready product teams can actually ship with.

Explore cap-x on GitHub

CaP-Gym

39 tasks across 3 simulators. Agents control robots by composing Python code.

CaP-Agent0

Training-free agents with visual differencing and auto-synthesized skill libraries.

CaP-RL

Post-training with GRPO reinforcement learning. Language models learn from environment rewards.

CaP-Bench

8-tier evaluation framework testing abstraction levels and interaction modes.

Pricing

Start free. Scale when you ship.

Integrate in under 1 hour. No ML expertise needed. Upgrade as your robot fleet grows.

Free

Always free
$0/month

Locomotion and balance skills that premium vendors charge thousands for — free.

  • Walking, running, jumping
  • Basic locomotion and balance
  • 1 robot platform
  • Community support

Pro

Save 33% annual
$99/robot/month

or $149/robot billed monthly

Task-specific skills for teams shipping real-world non-industrial deployments fast.

  • Everything in Free
  • 20 task-specific skills per year
  • Cloud simulation training
  • Up to 3 robot platforms
  • 1-hour integration support
  • Email support

Enterprise

Custom
From $2,500

/robot/month · min. 3-robot commitment

Industrial workflows, custom training, and dedicated support. Need a motion we don't have? We train it.

  • Everything in Pro
  • Unlimited skill deployments
  • Custom motion training on request
  • Warehouse, assembly, industrial workflows
  • Dedicated simulation environment
  • 99.9% uptime SLA
  • Dedicated success engineer
No credit card required to startTypical enterprise teams save $200k+ vs. building in-house per skillAnnual plans include priority onboarding

Your next robot skill

60 minutes from now,
your robot can do it.

No ML team. No simulation setup. No rebuilding from scratch.Just scan, train, ship — and stay ahead.

Explore cap-x on GitHub

No credit card required · Powered by our published research · Any robot platform