News · Research
Apptronik Apollo at GXO — warehouse pilot status
Source: GXO Logistics
Lead strip: Status: Research / early-stage pilot (GXO multi-phase R&D) · Availability: enterprise R&D only; public catalog — · All-in cues: capital strong; published fleet hours — · Scoring: methodology
Evidence brief only — not a sold ranking. Apptronik Apollo remains Research text-only on this site (no model-page push). Pilot announcements ≠ Digit-class paid fleet hours.
GXO Logistics has run more than one humanoid track. The densest public warehouse record still sits with Agility’s Digit program. Separately, GXO announced a multi-phase R&D / proof-of-concept with Apptronik around Apollo. This brief keeps those lanes separate, states what the Apollo × GXO materials actually say, and refuses to inflate a lab-to-DC roadmap into fleet evidence. For checkable product pages, use Digit and Figure 03.
Robot strips (context, not a trophy board)
- Digit · Agility Robotics · Status: Digit 4 field / Digit 5 announced · Use: warehouse logistics · Evidence: named sites + published hours/totes · → /robots/digit/
- Figure 03 · Figure AI · Status: Pilot (BMW Spartanburg sequencing) · Use: factory logistics · Evidence: BMW Group partner statement · → /robots/figure-03/
- Apptronik Apollo · Apptronik · Status: Research / early-stage GXO R&D · Use: warehouse PoC (company framing) · Evidence: GXO + Apptronik announcements; fleet hours — · text-only (no model page)
What GXO and Apptronik announced
On 20 June 2024, GXO announced an early-stage proof-of-concept program with Apptronik to develop practical warehouse applications for Apollo (GXO; Apptronik news mirror).
Documented framing from those materials:
- Multi-phase R&D: lab evaluation to fine-tune Apptronik’s AI model, then deployment to a U.S. distribution center once ready.
- Use-case language: labor-intensive DC processes — large movements (boxes/totes) through smaller ones (item pick, barcode scan) in CEO Jeff Cardenas’s comments.
- Hardware framing (vendor): Apollo stands about 5′8″, can carry ~55 lb, uses swappable batteries, linear actuators, force-control architecture, and a flexible safety-zone perimeter for work near people.
- GXO voice: Adrian Stoch (then Chief Automation Officer) described Apollo as having potential across the DC, including labor-intensive processes (The Robot Report).
What that confirms: a named 3PL and a named humanoid developer agreed to a structured PoC / R&D path aimed at warehouse work.
What that does not confirm: multi-year RaaS revenue, published tote counts, cumulative operational hours, robot fleet size at a named DC, or a public buy page for Apollo.
Pilot vs Digit-class fleet evidence
GXO also has a far denser public trail with Digit. Agility reports Digit 4 logging >65,000 operational hours across sites including GXO, plus 100,000+ totes at Flowery Branch under a multi-year commercial / RaaS agreement that followed a late-2023 pilot (Digit materials summarized on our Digit pages; primary: Agility Digit 5 unveil and 100k-totes post).
| Signal | Apptronik Apollo × GXO | Digit at GXO (Digit 4 ledger) |
|---|---|---|
| Announcement type | Early-stage PoC / multi-phase R&D (Jun 2024) | Multi-year commercial / RaaS (Jun 2024) after 2023 pilot |
| Published tote / hour KPIs | — in primary GXO/Apptronik releases | 100k+ totes; >65k hours (Agility, Digit 4) |
| Named live commercial structure | PoC / R&D | Multi-year RaaS language |
| Public model page on this site | Text-only (Research) | /robots/digit/ |
| Buyer takeaway | Track as Research / pilot | Score as field logistics evidence |
Do not merge Apollo’s PoC language into Digit’s tote-hour ledger, even though both involve GXO. Same customer name ≠ same evidence class. Criteria: methodology.
Factory-side contrast: Figure 03 carries a BMW partner statement for Spartanburg sequencing — again a different evidence lane (automotive logistics pilot, not Apollo warehouse KPIs).
Status reading for 2025–2026
Secondary registries and trade summaries describe Apollo progressing along GXO’s multi-phase path and remaining in an enterprise pilot / R&D posture into 2026, with some location attributions (e.g. Oklahoma DC language on third-party registries). Treat those as secondary until GXO or Apptronik publish primary KPIs for hours, totes, or robot counts.
Separately, Apptronik’s February 2026 financing (~$520 million Series A extension; valuation headlines in the ~$5–5.5B range with Google, Mercedes-Benz, and others) documents capital for commercialization and production ramp (Reuters; CNBC). CNBC also notes early Apollo versions working in designated areas with partners including Mercedes-Benz, GXO, and Jabil — still partner-zone / pilot language, not a published fleet dashboard.
Funding ≠ fleet. Capital and strategic partners improve continuity; they do not replace tote-hour receipts.
Why Apollo stays Research text-only here
Robotscomparison keeps Apptronik Apollo as Research text-only: no /robots/apollo/ push in this article, no invented Evidence Score, no trophy placement next to platforms with denser public operational ledgers. When readers need a warehouse product page with published hours/totes, start at Digit. When they need a BMW partner-stated factory sequencing trail, use Figure 03.
Unknowns that stay marked —:
- Apollo cumulative hours or totes at GXO
- Robot count in any named GXO DC
- Public list price / RaaS rate card
- Warranty / SLA for walk-up buyers
Practical takeaway
- Read Apollo × GXO as early-stage warehouse R&D, not as Digit-class commercial fleet proof.
- Keep GXO’s Digit hours/totes on the Digit ledger (Digit).
- Treat 2026 capital as continuity, not throughput (funding ≠ fleet).
- Cross-check factory pilot evidence on Figure 03 and scoring rules on methodology.
FAQ
Is Apollo commercially deployed at GXO like Digit?
Not on the same evidence class. GXO’s Apptronik program is framed as early-stage PoC / multi-phase R&D. Digit’s GXO trail includes multi-year commercial / RaaS language plus published tote and hour metrics on the Digit 4 ledger. See Digit.
Does Apptronik’s $520M raise prove warehouse readiness?
No. It proves investor demand and a commercialization / production-ramp budget. Published Apollo fleet hours at GXO remain — in the primary materials cited here.
Why is there no Apollo model page link?
This site keeps Apptronik Apollo Research text-only until public operational evidence and availability cues justify a model page. Navigation for denser evidence: Digit, Figure 03.
What should a warehouse buyer ask next?
Ask which generation and commercial structure would sit on the floor, for named site KPIs (hours, interventions, tote/cycle rates), and for a written safety case. If the answer is still “lab / PoC,” budget it as R&D — not as a Digit-class fleet replacement.
Sources
- GXO: Multi-phase R&D initiative with Apptronik (20 Jun 2024)
- Apptronik: GXO advances humanoid strategy
- The Robot Report: Apollo in tests by Apptronik and GXO
- Reuters: Apptronik raises $520M (11 Feb 2026)
- CNBC: Apptronik $520M / ~$5B valuation
- Robotscomparison: Digit, Figure 03, methodology
Apollo GXO hours, totes, robot count, and public list price: —.