AsapiThe AI Agent Trials
Asapi / Catalogue / Two boards placed together by a bimanual robot, one per hand
Environment datasheet · Robot manipulation

Two boards placed together by a bimanual robot, one per hand

ID llarms/task06 · revision 1 · results as of 2 November 2026

In a simulated robot workshop, the agent drives a two‑handed robot by typed commands so that each hand picks up its own board and sets it down, and both boards settle exactly inside their outlines together.

Checks
1, 1 point for the whole build, none for part of it
Tested
1 model, one attempt
Best
0% (GPT‑6.1 Sol)
Full solves
0 of 1
Attempt
12 min
What it tests

Two boards, one per hand, set down inside their outlines together

Robot control

Each hand carries one board to its outline, and both must be released in place within a set amount of simulated time.

How it works

A simulated workshop, two five‑fingered hands and a board for each

The agent's world

A coding agent sends JSON to a local robot server: read the state, or send a motion of 1 to 32 velocity segments.

The robot

Two seven‑joint manipulators on fixed bases, each with a five‑fingered hand that holds by friction alone, so boards can slip, turn or drop.

Limits

One session of 20 minutes of wall time, planning included, with action and simulation‑time budgets; no resets or direct state changes.

The reward

One point only if both boards settle together with every corner inside outlines of their own size, each lifted by its own hand.

Results

Both boards ended millimetres out when simulation time ran out

The dot is the one attempt.

MedianFully solved
1 modelOne attempt0%0 of 1
ModelReward
GPT‑6.1 SolOpenAI0%

A person can complete this build easily.

What models get wrong

It gets within millimetres, but its corrections never converge

Robot control

  1. Both boards came to rest outside their outlines, by 4.2 and 9.5 mm at the worst corner; in no saved state was either fully inside.

  2. Commands 13 to 31 alternated finger moves, wrist turns and pushes, many blocked by collisions, and releases and contact kept disturbing the alignment.

  3. After a second placement, regrasping and releasing both boards, two more commands used up the simulation‑time budget, with 8 of the 20 minutes left.

Running it

What an attempt takes, and what the environment includes

An attempt

Cost and tokens weren't recorded: the attempt ran on a subscription sign‑in, which logs neither.

What's in it

The frozen runtime with this scenario, the request and interface document, the grader and its self‑test, and the attempt's full record and videos.

Format

Node.js, with the physics in headless Edge and a Python grader; it runs on a Windows PC today, and other packaging is on request.

Fresh variants

Other builds and layouts in the same design, made on request and never published.

The files and grader aren't published; the job's row on The AI Agent Trials, our public benchmark, sums up the job and shows each model's score.

Try one of our three sample environments

Three environments are packaged as samples you can download, with their graders and reference solutions. Ask for them with your work email, or book a call about the rest of the catalogue.