Warrior Brief｜Robotics Q4 Leaderdrive / 绿的谐波 Web Page Package

Date: 2026-06-15 Owner: Hugo / Genius Team Research agent: Finance / Charlie AGT-002 Target worker: Warrior / Team Fullstack Status: READY_FOR_WARRIOR Visibility: PUBLIC Output: site Priority: P0 Target route: /robotics/supply-chain/leaderdrive/

0. Page title

Leaderdrive / 绿的谐波: Strong Supplier Signal, Not Yet CATL-or-NVIDIA Proof

Subtitle:

Filing-backed revenue, profit, and reducer-volume acceleration make Leaderdrive a real S4 supplier case; S5 still needs named humanoid customers, humanoid-specific revenue, durable value content, and margin proof.

1. Source files

Primary synthesis:

knowledge/robotics-leaderdrive-s4-to-s5-evidence-debt-v1.md

Supporting evidence:

knowledge/robotics-leaderdrive-morning-review-public-packaging-guardrail-v1.md
knowledge/robotics-leaderdrive-public-case-v2.md
knowledge/robotics-leaderdrive-case-study-slide-evidence-v3.md
knowledge/robotics-mainline-q0-q5-evidence-matrix-v1.md
knowledge/robotics-q1-value-stack-apple-android-or-something-else-v2.md

2. Page thesis

Leaderdrive / 绿的谐波 is a strong S4 filing-backed supply-chain signal, not yet a proven CATL-of-EV or NVIDIA-of-AI analogue for humanoid robotics.

The page should hold both ideas at once:

Leaderdrive matters because the operating evidence is real: revenue, profit, robot-component reporting line, reducer-unit sales, and Q1 follow-through are in filings.
Leaderdrive is not yet proven as a platform-defining humanoid supplier because public sources still lack named humanoid OEM design wins, contract value, humanoid-only revenue/margin, validated BOM attach, durable per-robot value content, and downstream customer economics.

Memorable frame:

Strong filings, open analogy.

3. Recommended page architecture

Section A — Hero

Core message:

This is not a concept-only supplier story. But it is also not yet a confirmed humanoid winner.

Use a two-column hero:

Why Leaderdrive matters:
- 2025 revenue / profit acceleration;
- robot / embodied-intelligence component reporting line;
- harmonic reducer volume acceleration;
- 2026 Q1 follow-through;
- product breadth beyond discrete reducers.
What is missing:
- named humanoid OEM;
- contract value / backlog / order volume;
- humanoid-specific revenue and gross margin;
- single-robot value content;
- validated BOM attach;
- margin durability under scale;
- downstream customer ROI/payback.

Section B — What is proven

Use cards with source grades.

Proven by filings 🟢:

2025 revenue RMB 570.714m, +47.31% YoY.
2025 attributable net profit RMB 124.367m, +121.42% YoY.
2025 “工业及具身智能机器人零部件” revenue RMB 422.528m, +52.61% YoY, gross margin 34.88%.
2025 harmonic reducer production 433,655 units, +72.27% YoY.
2025 harmonic reducer sales 425,158 units, +72.48% YoY.
2025 harmonic reducer inventory 18,238 units, -6.34% YoY.
2026 Q1 revenue RMB 140.134m, +42.96% YoY.
2026 Q1 attributable net profit RMB 32.634m, +61.17% YoY.

Message:

This is filing-backed supplier acceleration, stronger than theme exposure.

Section C — What is not proven

Use a “missing before S5” card grid.

Not proven before S5:

named humanoid OEM design win;
supply scope, contract period, order value, or order volume;
humanoid-only revenue split;
humanoid-specific gross margin;
single-robot value content;
validated production BOM attach;
customer ROI/payback downstream;
repeat deployment economics;
platform/module switching cost comparable to CATL or NVIDIA.

Message:

A supplier can be real before it becomes platform-defining.

Section D — CATL / NVIDIA analogy test

Do not ask “Is Leaderdrive the CATL/NVIDIA of robotics?” as a yes/no headline. Use a five-gate test.

Five gates:

Design-win gate
- Need named top humanoid OEM customer + supply scope.
- Current status: not disclosed in reviewed public sources.
Revenue gate
- Need humanoid-specific revenue / backlog / order volume separated from industrial robotics.
- Current status: filing line is industrial + embodied-intelligence robot components, not humanoid-only.
Architecture gate
- Need validated BOM showing reducer / actuator modules remain durable high-value content.
- Current status: product fit exists; platform attach is not proven.
Margin gate
- Need robot-related gross margin to remain durable while volume scales.
- Current status: 34.88% segment GM is a useful baseline, but not multi-period humanoid-specific proof.
Platform gate
- Need expansion from component to integrated actuator / joint module with qualification lock-in, switching cost, or service/testing ecosystem.
- Current status: product breadth is promising but not platform economics.

Section E — Product breadth / architecture hedge

Show product breadth as optionality, not proof.

Include:

harmonic reducers;
micro harmonic reducers;
rotary actuators;
frameless torque motors;
planetary roller screws / linear transmission;
planetary joint modules;
DD motors.

Message:

Product breadth can hedge architecture migration, but product display ≠ customer proof.

Section F — Anti-hype boundary

Include abnormal-trading caveat visibly.

Source anchor:

2026-06-04 to 2026-06-08 share price +36.00%.
Same period: SSE Composite -3.05%; automation-equipment index +2.13%.
Company said no undisclosed major matter.

Message:

Stock-price movement is market context, not order evidence.

Section G — Upgrade / downgrade dashboard

Upgrade toward S5 if:

named humanoid OEM design win appears with supply scope, volume, or contract period;
humanoid-specific revenue / backlog / margin is disclosed;
validated teardown / OEM BOM confirms production attach;
robot-related GM stays around 35% or improves while revenue grows above ~30% YoY;
integrated actuator / PRS / linear actuator revenue becomes material with attractive margin;
customer-side humanoid deployment economics validate repeatable downstream demand.

Downgrade if:

market narrative assumes named customers while filings remain anonymous;
revenue grows but gross margin declines materially;
inventory / receivables grow materially faster than revenue;
OEMs vertically integrate joints/actuators;
architecture shifts away from Leaderdrive's value-content areas;
repeated abnormal-trading disclosures deny undisclosed major matters while market assumes new orders.

Section H — Common misconceptions

Include these:

“RMB 422.528m is humanoid revenue.”
“Leaderdrive is confirmed supplier to Tesla / Figure / Unitree.”
“Harmonic reducers are the only humanoid value pool.”
“Share-price movement proves new orders.”
“Supplier revenue means CATL / NVIDIA analogy is already proven.”

Section I — Think Deeper questions

Use 4-6 questions:

If humanoid value migrates from discrete reducers to integrated actuators, does Leaderdrive gain value content or lose to OEM self-integration?
Is 34.88% robot-component gross margin sustainable under humanoid volume scale, or temporary mix/cycle benefit?
Which upgrade matters first: named OEM, humanoid revenue split, repeat order, validated BOM, or margin durability?
If low-cost humanoid OEMs compress ASP, do upstream suppliers win on volume or lose on price?
How should a public field guide present strong operating data plus hot price action without becoming promotional?

4. Must-include source anchors

绿的谐波 2025 annual report, disclosed 2026-04-23. PDF URL returned HTTP 200 on 2026-06-15; numeric anchors rely on previously extracted annual-report artifacts because the SSE URL returns an HTML/PDF wrapper in this environment. 🟢
绿的谐波 2026 Q1 report, disclosed 2026-04-23. PDF URL returned HTTP 200 on 2026-06-15; numeric anchors rely on previously extracted Q1-report artifacts. 🟢
绿的谐波 2026-06-09 abnormal-trading announcement. PDF URL returned HTTP 200 on 2026-06-15; numeric anchors rely on previously extracted announcement artifacts. 🟢
Leaderdrive official site: URL returned HTTP 200 on 2026-06-15. 🟢
Leaderdrive product page: URL returned HTTP 200 on 2026-06-15; keyword anchors product and harmonic matched. 🟢
Charlie CATL/NVIDIA five-gate analogy test and S4-to-S5 classification. 🟠

5. Do not say

Do not say 688017.SH is buy / sell / hold.
Do not include Hugo private portfolio weight, entry logic, or private rationale.
Do not infer Tesla / Figure / Unitree / AgiBot / UBTECH as customers without primary confirmation.
Do not call “工业及具身智能机器人零部件” pure humanoid revenue.
Do not treat share-price movement as customer/order evidence.
Do not use estimated reducer counts or BOM values as facts unless labeled 🟠 estimate with method.
Do not call Leaderdrive the CATL or NVIDIA of robotics as a conclusion.
Do not turn the case study into a valuation pitch.

6. Preferred wording

Use:

“S4 filing-backed supplier signal”
“strong filings, open analogy”
“industrial + embodied-intelligence robot components, not humanoid-only revenue”
“product fit ≠ customer proof”
“stock-price move ≠ order evidence”
“CATL/NVIDIA analogy test, not conclusion”
“component supplier vs module/platform supplier remains open”

Avoid:

“confirmed humanoid winner”
“pure humanoid revenue”
“supplies Tesla/Figure/Unitree” without primary proof
“order confirmed by stock move”
“CATL of robotics” as unsupported shorthand

7. Acceptance criteria

Warrior output is acceptable if:

The hero communicates both strong filing evidence and missing S5 proof.
The page separates revenue/profit, robot-component line, reducer volume, product breadth, and price-action caveat as distinct evidence units.
“工业及具身智能机器人零部件” is not labeled humanoid-only revenue.
No named customer is inferred without primary evidence.
CATL/NVIDIA is framed as a five-gate analogy test, not a conclusion.
The abnormal-trading caveat is visible.
No trade recommendation appears.
No Hugo private portfolio data or private rationale appears.

8. Public footer

Evidence map only. No trade recommendation. No private portfolio data. Product fit ≠ customer proof; stock-price move ≠ order evidence.