Screwed nine‑board grid with long rails by a bimanual robot
ID llarms/task08 · revision 1 · results as of 2 November 2026
In a simulated robot workshop, the agent drives a two‑handed robot by typed commands to lay two long rails flat and turn seven crosspieces across them, each exactly inside its outline, then fix all 14 overlaps with 42 screws, working from a stock fixture.
- 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
- 20 min
Two long rails and seven crosspieces laid as a grid, with 42 screws
Robot control
The robot carries two long rails and seven crosspieces from a stock fixture into a grid of 14 joints, fixed with 42 screws.
A simulated workshop, two hands, two drills and nine boards from a stock fixture
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 sliding 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 all nine boards settle inside outlines of their own size and all 14 joints are rigid with three screws each.
The long rails kept slipping, and no board ended inside its outline
The dot is the one attempt.
| Model | Reward |
|---|---|
| GPT‑6.1 SolOpenAI | 0% |
A person can complete this build easily.
It loses the long rails, collides with stock, and never reaches the drill
Robot control
Both hands took the first long rail and lost it in transport; the second, regrasped again and again, slipped while moving.
At commands 31 to 39 the manipulators collided with stock, and the attempt went back to recovering rails.
Three crosspieces were carried toward the grid, two of them turned; none ended inside its outline, the closest 42.7 mm out.
No drill trigger or engagement was ever sent: none of the 42 screws and none of the 14 joints at the 20‑minute limit.
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.
AsapiThe AI Agent Trials