X Square Robot Reports 1,816 Parcels an Hour in Live Test
X Square Robot says its stationary dual-arm system sorted 1,816 parcels in a one-hour, uncut August 12 livestream, with a reported success rate above 98%. The rate is about 45% higher than the 1,248-per-hour average from Figure AI's separate 200-hour demo, but the machines and tasks differ enough that the numbers should be treated as two company-run demonstrations, not a controlled head-to-head benchmark.
X Square Robot has published a 65-minute official recap of an August 12 logistics demonstration in which a stationary dual-arm system handled parcels continuously on a sorting line. The company labels the run a live challenge and reports 1,816 parcels in one hour at a success rate above 98%.
Pandaily separately reported the same counter total and described the stream as unedited. Those two retrieved records establish the event, but the performance figures remain company-reported: no independent technical audit, intervention log, or standardized benchmark result accompanied the public demonstration.
What the Figure AI comparison shows
The 1,248-parcel target shown for X Square's run closely matches the average from Figure AI's May endurance demonstration. Figure's robot team processed 249,560 packages across 200 hours, or 1,247.8 per hour. Against that arithmetic baseline, X Square's 1,816 figure is roughly 45% higher.
That comparison is useful context, not a controlled victory. X Square used a stationary dual-arm cell with simple grippers, while Figure's demonstration involved humanoid robots and a different operating setup. The public records do not establish identical parcel mixes, task definitions, error-counting rules, intervention policies, power constraints, or total system cost. Throughput alone therefore cannot establish which system would perform better in the same warehouse.
Why the longer public run matters
An uninterrupted hour is more informative than a short highlight reel because viewers can observe repeated cycles, recovery behavior, and variation over time. It still answers only part of a deployment decision. Logistics teams would need reproducible trials on their own parcel mix, documented failure and intervention rates, sustained uptime across full shifts, safety controls, integration requirements, and cost per successfully handled parcel.
The practical signal is that embodied-AI vendors are beginning to compete with longer, countable demonstrations rather than isolated clips. The next step is common evaluation conditions and independently reviewed operating data, so impressive counters can be translated into reliable production economics.
Key Points
- 1X Square Robot reports 1,816 parcels handled in one uninterrupted hour with a success rate above 98%.
- 2The reported rate is about 45% above Figure AI's separate 200-hour average, but different hardware and task conditions prevent a controlled comparison.
- 3Warehouse buyers still need intervention, uptime, safety, integration, and cost-per-success data from representative trials.
Scoring Rationale
A sustained public run offers more useful evidence than a short demo and targets a production-relevant logistics task, but company-reported metrics, non-equivalent systems, and missing intervention and cost data limit broader conclusions.
Sources
Primary source and supporting public references used for this report.
Practice with real Payments data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Payments problems
