Earnings call transcript: Leapmotor’s Q4 2025 showcases strong growth and profitability

Leapmotor reported FY2025 revenue of CNY 64.0 billion (+101.3% YoY) and turned profitable with net income of CNY 540 million; vehicle deliveries were 596,905 (+~103%) and gross margin improved to 14.5% (from 8.4%). Cash and equivalents rose to CNY 37.88 billion (+50.9%) and free cash flow reached CNY 7.8 billion; management projects EPS of $0.07 (FY2025) and $0.36 (FY2026) with revenue targets of $9.459bn (FY2025) and $13.853bn (FY2026). Company plans four major model launches in 2026 targeting 1.0m units and is expanding exports/localization, while citing risks from raw-material cost volatility, competition and regulatory/geopolitical headwinds.

Analysis

Leapmotor’s move to vertically integrate software-defined architectures and centralize domain controllers materially raises semiconductor content per vehicle and shifts aftermarket/service economics toward software and data. That creates a two-tier winners list: OEM-focused silicon vendors and data-center/edge infrastructure suppliers that capture recurring revenue from map/model updates and fleet telemetry, while commodity parts suppliers see margin pressure as OEMs internalize high-value subsystems.

The near-term risk is execution: rapid international rollouts and localization (Spain + Stellantis/FAW tie-ups) compress free cash flow and amplify exposure to logistics, tariffs and European regulatory timing over the next 6–18 months. A decisive regulatory clarification in the EU or a cell-localization announcement would be a binary catalyst; conversely, a sudden hardening of export rules or battery-cell shortages would reverse the upside quickly.

From a supply-chain vantage, Qualcomm-style ADAS content is the most direct lever—every incremental $ of silicon content scales cleanly with unit volume, unlike one-off marketing spends. The automotive push into large-model inference also creates measurable demand for short-latency on-prem/edge servers and rack-level AI appliances over a 9–24 month horizon—this is the structural angle that makes certain small-cap infrastructure names asymmetric beneficiaries.

Contrarian read: the market is overpaying for headline unit growth in OEM equities and underweighting suppliers that monetize per-vehicle software and cloud cycles. Near-term margin improvement claims are credible but fragile; prefer supplier/extractors of recurring software and compute dollars rather than the OEM equity run-rate until localization and regulatory tails are proven.

Key Decisions for Investors

• Long QCOM (12–18 months): buy QCOM stock or buy 12–18 month ATM calls to capture rising per-vehicle SoC content as OEMs adopt centralized architectures. Risk: mobile cyclical softness or weaker-than-expected automotive design wins; target asymmetric 2:1 upside vs downside with a 20–30% position sizing taper into regulatory milestones.
• Long SMCI (9–24 months): initiate a position in SMCI to play increased demand for edge and on‑prem inference hardware from automakers and Tier‑1s training/serving large driving models. Exit/trim on order-book disclosures from major OEMs or if gross margins compress >200bps; expect >50% upside if adoption accelerates.
• Buy MSCI on dips (6–12 months): accumulate MSCI exposure to capture flow upside from reweighting/ESG product demand as higher-rated Chinese EV issuers migrate into benchmarks. Treat this as a defensive play on product AUM growth; stop-loss if global passive flows reverse.
• Avoid direct long in Leapmotor equity until 2 catalysts are cleared (EU regulatory/tariff clarity and local-cell sourcing agreement) — instead express the thematic via semiconductor/infra suppliers to get exposure with less execution risk. Revisit OEM long only after positive confirmation of localization and sustained gross margins across two consecutive quarters.