Skip to main content
CUPEL
CUPEL
FIND YOUR WAY
Request a pilot

Find the footage worth training on.

Track record

Predicted first. Measured after.

Cupel records scores and predictions before collecting real-robot outcomes. The public track record shows reviewed quarterly aggregates from consenting customers.

0Published quarters
—Trials in latest cohort
—Customers in latest cohort
—Latest publication
What we publish

Three numbers, one per product.

Refine

Did the refined data help?

The gain in task success from refined hours against the same number of random hours, measured on the buyer's own policy, with a 95% interval and the trial count.

Test

Did the prediction hold?

How closely the world model's checkpoint ranking matched the ranking real robots measured later. Every prediction carries its world-model version.

Deploy

What happened on the fleet?

Success rate, interventions and late actions by anonymised deployment cohort.

Quarterly results

Refine pilot outcomes

No reviewed quarter has been published. A release requires outcomes from at least three consenting pilot customers.

Quarterly results

Test accuracy

No reviewed Test quarter has been published. A release requires verified rankable results from at least three consenting customers.

The gates

What has to be true, and when.

Each phase of Cupel ends with a test. We publish the result whether it passes or fails. Dates are estimates from November 2026.

GateWhenTestPass ifResult
AWeek 8Refined footage against random hours and simple filters on SO-101At equal hours, ≥10 pp over random and >0 pp over simple filtersFormal result pending
BWeek 16A paying customer reproduces the gain on their own training stack≥1 customer, ≥2 paid pilotsFormal result pending
CMonth 10World-model rankings against real-robot rankings, on held-out checkpointsRank correlation ≥0.8 across two robotsFormal result pending
DMonth 18Cupel's runtime against plain lerobot-rolloutFewer late actions and task success at least as highFormal result pending
The ledger

Why the numbers can't be fitted afterwards.

  • Written before the result. A prediction is stored before the real-robot trial runs, so it can't be adjusted to match.
  • Append-only. Rows go to a storage bucket with Object Lock. Each file records the hash of the file before it, so a deleted or changed row breaks the chain.
  • Versioned. Every row records the score version or world-model version that produced it.
  • Aggregates only. Scores, trial outcomes and fleet summaries. Never footage, checkpoints or recordings.