China's most closely watched automotive alliance just posted its weakest month of the year, and the explanation offered by its own brand communications is a masterclass in misdirection. HIMA, Huawei's Harmony Intelligent Mobility Alliance, delivered 37,490 vehicles in September 2026, down 29.15 percent from 52,916 a year earlier and 10.95 percent below August. It was the alliance's fourth consecutive year-on-year decline and its third consecutive month-on-month contraction, yet the accompanying scorecard chose to headline a cumulative total of more than 1.56 million vehicles and an unbroken delivery-speed record.
The scorecard also coincided with the most consequential governance change in the alliance's history. Two weeks before the September numbers were released, Huawei and Seres announced that Aito, the alliance's founding and highest-volume brand, would move to a new cooperation model with Seres taking the lead and Huawei providing support. On September 30 the two companies signed a new five-year agreement in Shenzhen to formalise it. Understanding why a brand that sells roughly half of HIMA's volume needed a structural reset, at the exact moment alliance volumes are falling, is the core analytical question for anyone with exposure to Chinese premium EVs.
The Scoreboard Doesn't Lie: 37,490 Units and a Fourth Straight Decline
HIMA's monthly trajectory in 2026 traces a clear arc: a strong first quarter built on the Aito M8 launch cycle, a plateau through the summer, and an accelerating decline as the year progressed. Year-to-date deliveries reached 367,229 units in the first nine months, up just 5.27 percent, a sharp deceleration from the 11.42 percent growth the alliance reported for the first eight months. Cumulative lifetime deliveries have now passed 1.56 million vehicles.
| Month (2026) | Deliveries | Same month 2025 | Year-on-year change |
|---|---|---|---|
| January | 57,915 | 34,989 | +65.5% |
| February | 28,212 | 21,517 | +31.1% |
| March | 26,582 | 22,912 | +16.0% |
| April | 32,759 | 27,555 | +18.9% |
| May | 46,124 | 44,448 | +3.8% |
| June | 50,624 | 52,585 | -3.7% |
| July | 45,422 | 47,375 | -4.1% |
| August | 42,101 | 44,561 | -5.5% |
| September | 37,490 | 52,916 | -29.2% |
The table exposes the mechanics of the decline. Growth turned negative in June and then deteriorated in a controlled, month-by-month fashion until September, when it fell off a cliff. The reason is base effect: September 2025 was the launch ramp of the Aito M8, and the fourth quarter of 2025 was the alliance's absolute peak, with 68,216 deliveries in October, 81,864 in November and 89,611 in December. Every 2026 comparison for the rest of the year will be measured against the most extreme quarter in HIMA's history.
More Models, Fewer Buyers: The Portfolio Dilution Math
The most revealing aspect of the September scorecard is what happened to the product range while volumes were falling. HIMA now spans five brands: Aito with Seres, Luxeed with Chery, Stelato with BAIC, Maextro with JAC, and Shangjie with SAIC. September saw the launch or pre-sale opening of the updated Luxeed R7, the Luxeed RX, the Stelato V8 and the updated Aito M8, whose pre-sales began on September 30 at a starting price of 389,800 yuan, some 30,000 yuan above the outgoing version.
Chinese market commentary has made much of the arithmetic: the alliance's addressable portfolio has expanded substantially year on year, from a handful of nameplates to a dense matrix of SUVs, sedans and now MPVs spread across five brand identities, yet total volume shrank. More products competing for the same showroom traffic and the same consumer consideration set is a textbook route to cannibalisation, particularly when several models share Huawei's ADS hardware stack, a common cabin architecture and overlapping price bands between 250,000 and 600,000 yuan.
That overlap is not incidental. HIMA's differentiation is largely software and brand storytelling layered on structurally similar vehicles, so incremental nameplates cannibalise each other before they expand the franchise. The alliance's own September explanation effectively conceded the point by promoting a delivery-speed record instead of a monthly comparison, a communications choice that obscures deterioration rather than addressing it.
The Aito Reset: Where the Money Goes, and Where It Doesn't
The September 15 restructuring moved operational control of Aito to Seres across product definition, product design, brand marketing, retail channels and service, with Huawei providing support. Reporting by Chinese outlets added the operational detail that matters commercially: dealer agreements for Aito switched contracting party from Huawei to a Seres affiliate, Seres took over settlement of comprehensive service fees for new orders from September 16, and parts of Huawei's outsourced Aito project teams were disbanded, with staff reassigned.
The financial backdrop explains the urgency. Seres purchased 56.05 billion yuan, roughly 8.32 billion US dollars, of goods and services from Huawei and its affiliates in 2025, about ten times the level three years earlier. Under the previous cooperation model, Seres paid Huawei technology licensing and channel marketing fees of approximately 10 percent of the price of every vehicle sold. Huawei's previous five-year agreement was approaching expiry, and negotiations had reportedly become tense.
- Huawei's position: retain technology licensing revenue and brand association while transferring operating cost, dealer liability and inventory risk to a partner that is now also the counterparty negotiating the fee schedule.
- Seres' position: gain control of its brand, its channel economics and its customer relationship before the fee structure is reset for another five years.
- The dedicated-network question: Aito is establishing an independent sales network, with some HIMA stores transferring to Seres to sell Aito exclusively, which converts a shared channel into a competitive boundary between allied brands.
- The user base: Aito's installed base has surpassed 1.2 million vehicles, an asset valuable enough that neither side can afford a messy separation.
Who Gains When HIMA Stalls
The beneficiaries of HIMA's slowdown are distributed across price bands. In the premium segment, Li Auto and a reinvigorated NIO continue to contest family SUVs, while Leapmotor and Zeekr set fresh delivery records in September, evidence that volume is rotating toward value-led and technology-led propositions rather than alliance-branded premium. Xiaomi's expansion into the same buyer demographics intensifies the contest for the 250,000 to 400,000 yuan band, precisely where Aito's volume models sit.
Maextro, the alliance's ultra-luxury brand, is a separate and more fragile story. Its V800 MPV is now the subject of a brake pedal bracket fracture controversy that has raised questions about validation rigour in the alliance's highest-margin products, a reputational headwind HIMA does not need while volumes decline. Western premium brands gain the most from the confusion: for every month that HIMA's own brands compete for the same customer, the German trio's long-established channel and residual-value narratives look comparatively reassuring to Chinese luxury buyers.
The Reality Check: Is Huawei Quietly Exiting the Driver's Seat?
Strip away the partnership language and a harder reading emerges. Huawei's consumer narrative has always insisted that it does not build cars, and that it sells intelligence and brand gravity to partners who own the factories, the dealers and the losses. The Aito reset is the most explicit expression of that doctrine to date. Huawei keeps the technology content and the marketing halo while Seres absorbs channel operations, dealer settlements and the commercial consequences of volume fluctuations. If Aito recovers, Huawei shares the credit; if it does not, Huawei's revenue per vehicle remains contractual and largely insulated from the showroom.
Two caveats keep this from being a clean win. First, Huawei's fee income depends on Aito's volume, so a 29 percent annual decline is a direct hit to licensing and channel revenue, not a hedge. Second, the brand-equity halo is not free-standing: it requires HIMA's flagship products to be visibly excellent. When the alliance's most expensive model suffers a safety-component controversy and its highest-volume brand's sales fall by nearly a third, the halo weakens at both ends of the portfolio simultaneously.
A third question is whether the decline is cyclical or structural. The 2025 comparison base was inflated by a launch cycle that will not repeat, and the 2026 sequence still shows total volume above 2024 levels. But the entry of China's 2026 purchase-tax framework, which halves the NEV exemption relative to previous years, removes a powerful demand subsidy precisely as premium prices rise. Aito's own M8 arrived with a 30,000 yuan price increase into a market where competitors are cutting. Price increases plus reduced tax support plus channel overlap is not a transition trough; it is a demand test.
Strategic Outlook: Three Scenarios for the Huawei-Seres Marriage
HIMA's future now hinges on whether dedicated brand operations restore momentum or simply make the alliance's internal competition visible to customers.
Bull Case
The dedicated Aito network lifts service quality and pricing discipline, the updated M8 and Luxeed portfolio convert pre-sales into volume, and HIMA finishes 2026 above 550,000 units. Seres proves that operational control plus Huawei technology is a better model than shared channels, and the template is extended to other brands in the alliance.
Base Case
Aito stabilises near 35,000 to 45,000 monthly units, the other four brands remain marginal, and HIMA ends the year flat to slightly positive. Huawei's licensing revenue holds, Seres absorbs dealer friction, and the alliance drifts from a growth story into a mature premium portfolio competing on refresh cadence.
Bear Case
Continued cannibalisation, a rising price war and renewed safety scrutiny push HIMA below 400,000 units for the year. Seres' margin absorbs the fee burden it just renegotiated, Maextro's credibility damage spreads to the wider Huawei brand, and partners such as Chery and SAIC quietly reprice their commitments to the alliance model.
Five considerations for investors and strategists:
- Read the scorecard, not the headline. Cumulative delivery totals and speed records are being used to mask a fourth consecutive annual decline in monthly volume.
- Track dealer contracting, not the press release. The switch of Aito dealer agreements and service-fee settlement to a Seres affiliate is the concrete evidence that operating control has genuinely moved.
- Model Huawei's revenue per vehicle, not unit volumes. Technology licensing and channel fees of roughly 10 percent of vehicle price make Huawei's economics dependent on premium mix rather than total deliveries.
- Watch the 2025 base effect dissolve in the fourth quarter. Comparisons against 81,864 and 89,611 monthly units will be the harshest test the alliance has faced, and the reaction will define its 2027 product strategy.
- Treat Maextro's quality dispute as a portfolio-wide risk signal. The V800 brake pedal bracket fracture case and the delivery decline are connected by a single variable: the credibility of Huawei's engineering governance across the alliance.