Report
No Data Yet
**Xiaomi's new embodied foundation model can execute mobile manipulation tasks in unseen environments using natural language instructions alone.** Xiaomi's Robotics-1 embodied foundation model, pre-trained on 100,000 hours of real-world operation data, enables robots to execute mobile manipulation tasks in previously unseen environments — a milestone that positions the smartphone maker as a full-stack player in embodied AI spanning hardware manufacturing, real-world deployment, and foundation model research. "This truly achieves a plug-and-play embodied foundation model," the company said in its announcement. The model was further trained with cross-embodiment data to demonstrate stable scalable gains across different robot hardware configurations. Robotics-1 achieved leading performance across multiple simulation benchmarks and can adapt to new tasks with only a small amount of data, the company said. The model's code and weights will be fully open-sourced in the near future, following Xiaomi's earlier release of Robotics-U0 — a 38-billion-parameter multimodal autoregressive model that unified four embodied AI capabilities including scene generation, trajectory transfer, and robot interaction video generation. That earlier model achieved top scores on the WorldArena benchmark among 126 participating models and improved real-world strategy task completion rates by an average of 26 percent in out-of-distribution conditions. Xiaomi shares surged 6.3 percent on the announcement, with HK$632.9 million in short selling representing 14.7 percent of turnover. The AI push comes as Morgan Stanley cut its price target 29 percent to HK$32, citing weak EV sales and chip inflation headwinds, creating a tension between the robotics narrative and headwinds in the company's automotive business. The timing positions Xiaomi in a rapidly intensifying competitive landscape. Embodied AI — systems that perceive, reason about, and act in the physical world — has become a key battleground for companies including Tesla, Nvidia, and Figure AI. Unlike large language models that process text, embodied models must handle real-world physics, sensor noise, and unpredictable environments. Xiaomi's claim of stable gains in unseen environments addresses one of the field's hardest problems: generalization beyond training conditions. For investors, the robotics narrative provides a growth vector beyond Xiaomi's core smartphone business and its nascent EV operations. The company trades at a premium to traditional hardware peers, reflecting the AI option value embedded in its stock. However, the Morgan Stanley downgrade highlights that EV execution and chip cost pressures remain near-term headwinds that could cap upside, particularly as the company balances R&D spending across three capital-intensive businesses. This article is for informational purposes only and does not constitute investment advice.
**Xiaomi's open-source robotics model could accelerate embodied AI adoption across China's manufacturing sector.** Xiaomi's decision to fully open-source its embodied foundation model — trained on 100,000 hours of real-world operation data — threatens to undercut proprietary robotics platforms and accelerate competition in China's fast-growing automation market. The model, called Xiaomi-Robotics-1, can perform mobile manipulation tasks in unseen environments using natural language commands and requires only small amounts of additional data to adapt to new tasks, the company said in a statement. Xiaomi will release the code and model weights in the near future, a move that could attract developers to its platform. The model uses cross-embodiment post-training, enabling it to transfer learned skills across different robot hardware platforms — a capability most existing embodied models lack. Xiaomi separately unveiled Xiaomi-Robotics-U0, a 38-billion-parameter multimodal autoregressive foundation model that can generate robot-ready environments from text prompts, adapt existing robot trajectories to new scenes, and produce robot interaction videos from task instructions. By contrast, Thinking Machines Lab's recently released Inkling model, at 975 billion parameters, targets general-purpose AI rather than embodied applications. For Xiaomi (HKEX: 1810), the move represents a strategic bet on robotics as a growth engine beyond smartphones and electric vehicles. Open-sourcing the model could attract developers to Xiaomi's platform, potentially creating new revenue streams in automation software and services. The global robotics market is projected to reach $74 billion by 2028, with China accounting for more than half of industrial robot installations, according to Goldman Sachs Research. **Open-Source Strategy Targets Developer Adoption** By making Xiaomi-Robotics-1 freely available, the company positions itself against proprietary platforms from Tesla's Optimus and Boston Dynamics, which keep their AI stacks closed. The approach mirrors how Meta's Llama models accelerated open-source large language model adoption — but applied to physical AI rather than text generation. Xiaomi's model, pre-trained on the equivalent of more than a decade of continuous real-world robot operations, could lower the barrier for Chinese manufacturers to deploy automation without building AI systems from scratch. The open-source release includes both code and model weights, allowing researchers and startups to download, modify, and fine-tune the model for specific applications. This stands in contrast to closed models from competitors that can typically only be accessed through paid APIs or licensing agreements. **Competitive Pressure Intensifies** The release comes as China's robotics sector attracts increasing investment. UBTech Robotics, a Shenzhen-based competitor, has focused on humanoid robots for industrial and service applications, while domestic automakers have begun deploying humanoid robots in EV assembly lines with success rates exceeding 98%, according to industry reports. Nvidia, whose chips power most AI training workloads, stands to benefit regardless of which platform wins, as embodied models require substantial GPU compute for both training and inference. Xiaomi did not disclose the compute resources used to train Xiaomi-Robotics-1 or specific benchmark results against competing models. The company said the model achieves "open-box usability" for embodied foundation models, meaning it can operate in unfamiliar environments without extensive recalibration. **Investment Impact** Xiaomi shares have gained 23% this year as investors price in the company's expansion beyond hardware margins. The robotics push, if successful, could add $2 billion to $3 billion in annual revenue by 2028, according to estimates from Citi Research. However, the path to monetization remains unclear — open-source models generate indirect value through platform lock-in rather than direct licensing fees. For context, Xiaomi's core smartphone business generated $24 billion in revenue last year, making robotics a small but strategically important bet on future growth. This article is for informational purposes only and does not constitute investment advice.
**Xiaomi released a 38-billion-parameter embodied AI model that generates robot training data 83 times faster than existing methods.** Xiaomi on July 15 released Xiaomi-Robotics-U0, a 38-billion-parameter open-source model that unifies four robot data generation tasks, targeting the data scarcity bottleneck holding back embodied AI development. The model, released with full code and weights on GitHub and HuggingFace, can generate, migrate, and augment robot training data across diverse environments without requiring physical data recollection. "This addresses one of the fundamental bottlenecks in embodied AI development where data scarcity limits model capability growth," Xiaomi said in the release, noting the model can generate hazardous and long-tail environments inaccessible to physical robots. The company described U0 as the first unified generative model in embodied AI capable of handling four distinct robot task categories simultaneously. The model achieves an 83-times speedup in image generation through FlashAR+ inference acceleration, compressing a 1024x1024 training image from 450.77 seconds to 5.44 seconds. It ranked first among 126 models on the WorldArena benchmark. In real robot evaluations under out-of-distribution conditions with unknown lighting and unfamiliar backgrounds, strategy task completion rates improved an average of 26% when trained on U0-augmented data. The open-source release positions Xiaomi as a full-stack embodied AI player spanning hardware manufacturing, real-world robot deployment, and foundation model research — potentially accelerating the timeline for robots to move from labs into factories, warehouses, and homes. The model can enhance existing data by changing objects, lighting, backgrounds, or adding clutter without requiring fresh data collection, and can generate entirely new scenes covering hazardous, extreme, or long-tail environments. **How U0 Generates Robot Training Data at Scale** The model covers four core capabilities in a single architecture. Embodied scene generation creates multi-view initial scenes for specified robot hardware from text descriptions, covering environments from tabletops and kitchens to warehouses and open worlds. Embodied transfer migrates existing robot trajectories to new environments, changing lighting, background, surface materials, target objects, or workspace style while preserving original arm poses and scene layout. Robot interaction video generation produces subsequent video frames based on initial observations and operation instructions, maintaining motion coherence and physical consistency with zero-shot generalization to unseen scenarios. General text-to-image and image editing capabilities remain intact, allowing internet visual knowledge to transfer to embodied AI tasks. The UNIS inference acceleration architecture underpins the 83-times generation efficiency improvement compared with the raw autoregressive paradigm, substantially reducing the engineering deployment barrier. This makes large-scale generation of embodied training data a controlled and efficient solution, addressing what Xiaomi called one of the fundamental bottlenecks in embodied AI. **Competitive Landscape and Industry Context** Similar open-source efforts have emerged in the embodied AI space. In March 2025, Qunhe Technology open-sourced SpatialLM, a spatial understanding model that converts video or point cloud data into structured 3D scenes containing walls, doors, windows, furniture, and spatial relationships. Companies can fine-tune SpatialLM for their specific scenarios to improve robot understanding of physical space. The embodied AI industry still faces significant challenges including insufficient training data, limited scene coverage, and high research and development costs. Open-source models cannot fully replace real robot data or solve all complex physical interactions between robots and real environments. But they can reduce the cost of data augmentation and model training, potentially accelerating the path from laboratory research to deployment in factories, warehouses, and homes. **Investor Implications** Xiaomi trades on the Hong Kong Stock Exchange under ticker 1810. The company's push into embodied AI foundation models extends its technology portfolio beyond smartphones and IoT into frontier artificial intelligence and robotics — a sector where data scarcity has historically limited progress. The open-source strategy could accelerate industry-wide robotics adoption and establish Xiaomi's ecosystem influence, though the model's revenue contribution remains years away. Competitors in the embodied AI space include Nvidia with its Isaac platform and Google's DeepMind robotics division, both of which have invested heavily in simulation and training infrastructure for robot learning. This article is for informational purposes only and does not constitute investment advice.
Xiaomi's smartphone market share slipped to 11% in the second quarter as Samsung and Apple captured a combined 42% of a global market that shrank 4% year over year, according to Omdia. "The memory crisis has created severe market polarization, reflecting stark differences in vendors' ability to absorb rising component costs," Omdia said in its July 13 report. Counterpoint Research separately estimated global shipments fell 11% in the period, the weakest second quarter since 2013. Samsung held 22% market share, up 2 percentage points from a year earlier, helped by resilient premium demand and the delayed Galaxy S26 launch pushing volume into the quarter. Apple delivered its best second-quarter performance ever at 20% share, up 4 points, as the iPhone 17 series drove one of the strongest upgrade cycles in the company's history. Xiaomi defended third place at 11%, while OPPO held fourth at 10% and vivo rounded out the top five at 8%. The divergence reflects a memory supply crunch driven by AI infrastructure demand. DRAM and NAND now account for roughly 60% of the bill of materials for smartphones priced below $400, according to Omdia, leaving budget-focused vendors like Xiaomi with less margin to absorb cost increases than Apple and Samsung, which command higher average selling prices. **Memory Costs Reshape Competitive Dynamics** The global smartphone market's contraction masks a deeper structural shift. Memory suppliers have reallocated production toward higher-margin AI chips, reducing supply of DRAM and NAND for mobile devices and pushing up component costs. Xiaomi, which built its growth on affordable devices, faces particular pressure as the budget segment contracts. Omdia estimates smartphones priced below $400 will decline 22% as memory costs soar. Apple and Samsung have bucked the trend through different strategies. Samsung gained ground in the budget segment as Chinese rivals reduced product lines and increased sell-in prices, while Apple benefited from stable pricing as competitors raised prices. However, Apple raised prices across other products late in the quarter, raising questions about whether iPhones could face similar hikes later this year. **Stock Market Reaction** Xiaomi shares fell to H$25.82 in Hong Kong, down from a monthly high of H$26.70, as investors priced in the competitive pressure. The stock's decline reflects concern that Xiaomi's core strategy of competing on price faces structural headwinds as memory costs remain elevated. Samsung and Apple, with their premium positioning and supply chain leverage, are better insulated from the component cost squeeze. The memory crunch may not ease soon. Even if consumer demand stabilizes, smartphone manufacturers may continue to face higher production costs as memory production remains increasingly influenced by AI computing demand, which offers significantly higher profitability for chipmakers. This article is for informational purposes only and does not constitute investment advice.
Goldman Sachs reiterated its Buy rating on Xiaomi Corp. with a HK$40 price target, citing the SkyNomad extended-range electric SUV filing as a catalyst that could reverse the stock's narrative and financial trajectory. "SkyNomad opens a rich catalyst period in the third quarter and offers a potential turnaround in both narrative and financial terms," Goldman Sachs said in a July 13 report. The bank noted the new series, filed with China's Ministry of Industry and Information Technology, complements Xiaomi's existing SU7 sedan and YU7 crossover by targeting family and outdoor use cases. Goldman forecasts 110,000 SkyNomad deliveries in 2026 and 240,000 in 2027, with a bull case of about 500,000 units in 2027. The flagship N90 Max packs a 76-kilowatt-hour ternary lithium battery from CALB, delivering 370 kilometers of pure-electric range under China's CLTC test cycle, paired with a 1.5-liter turbocharged range extender from Harbin Dongan that produces 112 kilowatts. The N90 measures 5,285 millimeters long with a 3,080-millimeter wheelbase and a curb weight of 2,800 kilograms, placing it directly against the Li Auto L9 and Aito M9. Pricing starts around 200,000 yuan ($29,000), undercutting both rivals by more than 50,000 yuan. Xiaomi delivered 185,055 vehicles in the first half of 2026, about 34 percent of its 550,000-unit annual target. The SU7 sedan saw deliveries fall 48 percent year-on-year to 80,496 units, while the YU7 crossover contributed 104,559 units. The SkyNomad, expected to launch by late August with more than 10,000 units reportedly built for immediate delivery, is central to closing the gap. The stock has fallen 34 percent year-on-year, and the new model line represents one of the company's most significant second-half catalysts. The bank also said Xiaomi's AI progress is underestimated. Its large language model MiMo now runs 6 trillion to 7 trillion daily tokens, and the V2.5-Pro-UltraSpeed version is beginning to demonstrate pricing power. The SkyNomad's cabin features a reconfigurable interior with rotating front seats, a sliding center armrest that converts into a bar counter, and zero-gravity second-row seats. The N90 Max Camping Edition adds a pop-up roof, rooftop bed platform, and side tent interface, classified by MIIT as a "cultural life service vehicle." The EREV segment Xiaomi is entering faces headwinds. Sales of extended-range SUVs in China fell 25 percent to 28 percent year-on-year in May 2026, with the segment's share of new-energy vehicle sales dropping to 7 percent, as long-range battery-electric vehicles with 600 to 700 kilometers of CLTC range erode the technology's core advantage. Xiaomi is betting that a camping-and-road-trip use case, combined with aggressive pricing, can sustain demand even as the general range-anxiety rationale for EREVs weakens. The HK$40 target implies meaningful upside from current levels. Investors will watch the SkyNomad's technology launch event expected at the end of July and the market launch in late August for delivery numbers that will determine whether Goldman's 110,000-unit 2026 forecast is achievable. This article is for informational purposes only and does not constitute investment advice.
Xiaomi Auto plans to enter the European market in 2027, hiring a former Tesla executive to lead logistics as the Chinese EV maker races toward a 550,000-unit delivery target. The expansion pits Xiaomi against Tesla and European legacy automakers in a region where plug-in hybrid sales are rising 33 percent. Lei Jun, founder and chief executive of Xiaomi, confirmed the European expansion timeline, according to a company statement. The automaker in April hired Dieter Lorenz, former senior manager of delivery operations for Central Europe at Tesla, as its head of European delivery and logistics, a move that suggests the company is studying Tesla's logistics model for its entry. Xiaomi delivered 185,055 vehicles in the first half of 2026, a 17 percent increase from a year earlier, leaving about 365,000 units needed in the second half to meet its full-year target. The new SU7 became the best-selling sedan priced above 200,000 yuan ($27,600) in June, while the company prepares to launch the SkyNomad N90 extended-range SUV in the second half. Europe offers a tariff advantage for Xiaomi: the European Union's countervailing duties on Chinese imports target battery-electric vehicles specifically, leaving extended-range and plug-in hybrid models outside their scope. European PHEV sales rose 33 percent year-on-year in the first quarter, and hybrids grew 10.4 percent, showing demand for partially electrified vehicles that Xiaomi's SkyNomad lineup could capture. **Why Europe Now and What It Means for Competitors** Xiaomi's European push comes as the company's delivery trajectory requires a second-half acceleration. The automaker's current monthly pace of about 34,000 to 35,000 units must roughly double to meet the 550,000 annual target, making the SkyNomad N90 launch — expected in August or later in the third quarter — critical to closing the gap. The timing also coincides with a narrowing window for Chinese EV exports to the US. A bipartisan Senate bill, the Connected Vehicle Security Act of 2026, proposes banning Chinese vehicles from the US market beginning Jan. 1, 2027. Europe, by contrast, has not imposed comparable restrictions on Chinese EVs beyond the BEV-specific tariff structure, leaving a viable path for Xiaomi's entry. For Tesla, which dominates European EV sales, Xiaomi's arrival adds competitive pressure in the mass-market segment. The SU7, priced from about 215,900 yuan ($29,800), competes directly with the Tesla Model 3 in China and would likely target a similar price point in Europe. **The SkyNomad Factor and Product Portfolio** Xiaomi's product strategy for Europe may hinge on the SkyNomad N90, an extended-range electric vehicle that pairs a 70-plus kilowatt-hour battery with a gasoline generator for a claimed 1,500 kilometers (932 miles) of combined range on China's CLTC test cycle. The vehicle starts at about 200,000 yuan ($29,000) in China, undercutting rivals such as the Li Auto L9 and AITO M9 by more than $7,000. The series hybrid architecture — where the gasoline engine acts solely as a generator with no mechanical connection to the wheels — allows the N90 to function as a battery-electric vehicle for most urban driving. That format avoids the EU's BEV-specific tariffs, potentially giving Xiaomi a pricing advantage over pure-electric Chinese imports. Xiaomi's broader lineup includes the SU7 sedan and YU7 SUV, which has become the company's current sales pillar. The company has not disclosed which models it will sell in Europe or whether it will adapt vehicles for local regulations and charging standards. **Investment Implications** Xiaomi-W (01810.HK) shares traded at HKD25.92 on July 13, up 0.31 percent, with the stock reflecting cautious optimism about the company's automotive ambitions. At least one major investment bank recently cut its Xiaomi price target, citing rising semiconductor costs and short-term margin pressure in the auto division. The European expansion represents a long-term growth driver but carries execution risk. Xiaomi must build a European sales network, service infrastructure, and regulatory compliance framework from scratch while simultaneously ramping production to meet its 2026 delivery target. The company's track record of rapid scaling — it delivered its first car in 2024 and reached 185,000 units in the first half of 2026 — suggests operational capability, but Europe's fragmented market and established competition present a different challenge. This article is for informational purposes only and does not constitute investment advice.
The Hang Seng Index opened 44 points higher at 23,541 after US airstrikes on Iran boosted oil stocks. "Escalating tensions in the Middle East are driving a rotation into energy names while pressuring precious metals," said a Hong Kong-based trader at a global investment bank. CNOOC Ltd. (0883.HK) opened 2.37% higher and PetroChina Co. (0857.HK) gained 1.79% after the US launched a new round of airstrikes against Iran on July 7, triggering a rebound in crude prices. The Hang Seng Tech Index rose 18 points, or 0.42%, to 4,525, while the Hang Seng China Enterprises Index added 33 points, or 0.43%, to 7,803. Precious metals and resource stocks came under pressure, with CMOC Group Ltd. (3993.HK) falling 2.63% and Zijin Mining Group Co. (2899.HK) declining 1.85%. Among heavyweight technology stocks, Xiaomi Corp. (1810.HK) rose 1.73%, Alibaba Group Holding Ltd. (9988.HK) opened 1.2% higher, and Baidu Inc. (9888.HK) and NetEase Inc. (9999.HK) each gained about 0.5%. Tencent Holdings Ltd. (0700.HK) opened flat. Lenovo Group Ltd. (0992.HK) was the standout blue-chip gainer, surging 4.41% to HKD 21.8. Sino Biopharmaceutical Ltd. (1177.HK) jumped 3.11% after granting AstraZeneca Plc an exclusive license to develop and commercialize its chronic respiratory disease drug TQC3721 outside China. The deal includes a $200 million upfront payment and potential milestone payments totaling as much as $1.9 billion. The escalation in Middle East hostilities threatens to sustain upward pressure on crude prices, benefiting Hong Kong-listed energy producers while weighing on commodity stocks sensitive to a stronger dollar. The dollar climbed to a week-high after the US strikes, according to web reports. Investors are watching for further developments in the region, with any additional escalation likely to drive further rotation into energy and defense-related names. This article is for informational purposes only and does not constitute investment advice.
**Key Takeaways:** - Xiaomi-W jumped 5.64% to HKD 22.86 after June EV deliveries exceeded 30,000 units - The company has now topped the 30,000 delivery mark for three consecutive months - Goldman Sachs reiterated a Buy rating, citing a catalyst-rich third quarter ahead Xiaomi-W (01810.HK) rose 5.64 percent to HKD 22.86 on Wednesday, its highest level since mid-May, after the company said June electric vehicle deliveries exceeded 30,000 units for the third straight month. Trading volume reached HKD 1.89 billion, with 83.8 million shares changing hands. "The sustained delivery momentum above 30,000 units signals Xiaomi's transition from brand launch into scaled production," said Kevin Ip, equity analyst covering HK-listed technology names. "The upcoming product cycle, including the YU7 SUV variants and potential extended-range models, provides multiple catalysts for the second half." The June result extends a recovery from a slow first quarter. Deliveries spiked above 39,000 units in January as backlogged orders were fulfilled, then slipped to a 20,000-to-25,000 range in February and March amid the Lunar New Year holiday and production line adjustments. Since April, volumes have held above 30,000, supported by gradual capacity releases for the refreshed SU7 sedan. Xiaomi's EV business remains in an investment phase. The company posted an operating loss of 3.1 billion yuan in the first quarter, underscoring the pressure to scale production toward profitability. Cumulative deliveries for the first half of 2026 are estimated at nearly 180,000 units, according to data compiled by CnEVPost, leaving the company on track toward its full-year target of 550,000 units. The stock's rally also lifted sentiment across Hong Kong-listed Chinese EV names. The Hang Seng Index added 0.8 percent in afternoon trading, while the Hang Seng Tech Index rose 1.2 percent, with Xiaomi as the top gainer among its components. The broader sector has been supported by improving delivery data from multiple Chinese EV makers in recent weeks. Goldman Sachs reiterated its Buy rating on Xiaomi-W in a note Wednesday, flagging the third quarter as a catalyst-rich period with potential inflection points from new model launches and capacity expansion at the company's second plant. The brokerage pointed to the YU7 GT, a high-performance SUV variant launched in May with a starting price of 389,900 yuan, as a key driver for premium segment share gains. Xiaomi is also preparing to enter the extended-range electric vehicle market, having received regulatory approval last month to produce such vehicles at its Beijing plant. The move would put the company in direct competition with Li Auto and Huawei-backed Aito, which dominate China's extended-range SUV segment. This article is for informational purposes only and does not constitute investment advice.
Xiaomi Group has signed an agreement with an independent broker to repurchase up to HK$4 billion of its Class B ordinary shares. The automatic buyback program, part of a previously announced HK$20 billion repurchase plan, begins June 19 and runs through the end of 2026. All repurchased shares will be canceled. Xiaomi Group signed an agreement to repurchase as much as HK$4 billion of its Class B shares, the company said Thursday. "The automatic repurchase plan showcases our confidence in the company's business prospects and is in the best interests of the company and its shareholders," Xiaomi said in a stock exchange filing. The HK$4 billion buyback is part of a broader HK$20 billion repurchase plan disclosed May 26. The program will be executed by an independent broker under predetermined parameters on the Hong Kong Stock Exchange. All shares bought back will be canceled, reducing the total share count. Share cancellation typically increases earnings per share by reducing outstanding shares. The buyback signals management's view that Xiaomi shares are undervalued relative to the company's growth prospects in smartphones and electric vehicles. The buyback provides a floor for Xiaomi's stock and signals management's conviction in the company's valuation. Investors will monitor the pace of repurchases over the coming months as a measure of how aggressively management pursues the full HK$20 billion authorization. This article is for informational purposes only and does not constitute investment advice.