Skip to content
Attune

Attune

Loading

Attune

The open frontier for adaptive robot intelligence.

Compete to build the next generation of robot models toward real-world physical control.

Explore Vector →Explore Horizon →Read the Docs

02The direction

AdaptiveIntelligence.More tasks.More robots.

A robot should not need a new brain every time the work changes.

03Technology

Compete at the edge of Physical AI.

  1. 01
    In-Context Robot Learning+ Human-Video Learning
  2. 02
    World Action Models+ WAM Inference Optimization

01 / Action-aware in-context learning

Vector

02 / Human-video ICL × World Action Models

Horizon

01 / Action-aware in-context learning

Vector

Learn the behavior. Adapt the execution.

Vector pushes adaptive manipulation from explicit robot demonstrations toward greater precision, robustness, and transfer.

Input

  • Demonstration video
  • Robot action trajectory
  • Current robot observation

VECTOR

  • Robot actions

What we measure

  • Adaptation

    Execute demonstrated behavior under changed conditions.

  • Precision

    Solve physical tasks where small errors matter.

  • Robustness

    Handle changes in position, environment, objects, and clutter.

  • Cross-Embodiment

    Move toward reusable intelligence across robot systems.

Our direction

  1. Spatial Variation
  2. Environment Variation
  3. Object Variation
  4. Clutter
  5. Cross-Robot

02 / Human-video ICL × World Action Models

Horizon

Watch the task. Model the future. Act.

Horizon brings several of Physical AI's newest directions into one competition: human-video task prompting, in-context adaptation, World Action Models, and efficient inference.

Input

  • Human / robot demonstration video
  • Robot history

HORIZON

  • Future physical state
  • Robot actions

What we measure

  • All in Vector

    Adaptation · Precision · Robustness · Cross-Embodiment

  • Human-Video Learning

    Teach unfamiliar physical work through natural visual demonstrations.

  • Real-Time WAM

    Preserve frontier capability while pushing inference toward practical closed-loop control.

Our direction

Performance OptimizationSpeed Optimization↻ iterate
SOTA

Join our open competition

Attune is built around measurable & auditable competition.

  1. 01Same Conditions

    Compared models face the same task, demonstration, starting state, seed, and evaluation rules.

  2. 02Randomized Evaluation

    Objects, environments, appearance, and clutter change so models must generalize rather than memorize.

  3. 03Reproducible Results

    Important frontier results should be independently verifiable.

Beat today's frontier. Define the next one.

Start with Vector →Start with Horizon →Read the Docs