Unitree vs UBTech: Which Humanoid Robot Company Is Stronger?

 

With Unitree opening subscriptions on 10 August and a STAR Market debut expected in the second half of the month, humanoid robotics finally gets two directly comparable listed companies. On one side is Hong Kong-listed UBTech, public since late 2023 and long branded the first humanoid robotics stock, which posted 2025 revenue of 2.001 billion yuan, up 53.3%, against a net loss of 790 million yuan and cumulative losses exceeding 3 billion yuan over three years. On the other is Unitree, arriving on the A-share market with 2025 revenue of 1.708 billion yuan, up 335%, a 60.27% gross margin and adjusted net profit of 600.1 million yuan, profitable every year since 2020. Two of the most watched Chinese companies in the sector, one expanding through losses and one decelerating in profit. Understanding the divergence between these paths matters more than debating whose robot looks more impressive, because they embody two fundamentally different assumptions about how humanoid robotics gets commercialized.

 

 

 

Unitree posted 2025 revenue of 1.708 billion yuan, up 335%, with a 60.27% gross margin and adjusted net profit of 600.1 million yuan, having stayed profitable since 2020, a rarity in the sector.

 

UBTech posted 2025 revenue of 2.001 billion yuan, up 53.29%, with full-size embodied humanoid revenue of 821 million yuan surging 2,203.7% and gross margin improving to 37.7%, but a net loss of 790 million yuan and over 3 billion yuan in cumulative losses across three years.

 

Per Omdia data, the top three humanoid shipment leaders in 2025 were Agibot at 5,168 units, Unitree at 4,200 and UBTech at 1,000, leaving UBTech with roughly 7.5% share, well behind the leaders.

 

The strategies differ sharply, with Unitree pursuing volume through affordability as average humanoid prices fell from 590,000 to 166,400 yuan, while UBTech targets large industrial and commercial contracts, winning a single order near 100 million yuan in July.

 

Unitree's growth is decelerating, with first-half 2026 revenue guided up roughly 35.6% to 45.4% against 332% a year earlier and adjusted profit expected to decline about 6% to 22%, while UBTech narrows losses toward a 2026 target of 10,000 units in industrial humanoid capacity.

 

 

The fastest way into this comparison is to place the two income statements side by side.

 

Per 36Kr's analysis of the prospectus, Unitree posted 2025 revenue of 1.708 billion yuan, up 335%, with adjusted net profit of 600 million yuan, up 674%, and a core business gross margin of 60.27%. In the same year, UBTech generated 2 billion yuan in revenue with a loss of 790 million yuan. UBTech is larger by revenue, but the two are not in the same league on earnings quality.

 

 

UBTech's numbers are not weak in themselves. Per TradingView's summary of its annual report, 2025 revenue grew 53.3% to 2.001 billion yuan led by a 2,203.7% surge in industrial humanoid sales, gross margin improved from 28.7% to 37.7%, and the net loss narrowed to 789.8 million yuan. Per BigGo Finance's review, full-size embodied humanoid revenue reached 821 million yuan.

 

The question is sustainability. Revenue climbed from 1.047 billion yuan in 2023 to 1.295 billion in 2024 and 1.981 billion in 2025 on one accounting basis, while losses narrowed from 1.234 billion to 703 million yuan, yet cumulative three-year losses still exceed 3 billion yuan. Founder Zhou Jian has described the company as walking on three legs, splitting energy evenly between commercial and industrial applications and home use, a multi-front strategy that carries a heavy capital efficiency cost.

 

 

Unitree took the opposite route, building cost advantage and scale in quadrupeds before migrating those capabilities to humanoids. Per Tech Market Briefs' prospectus review, humanoids rose from 27.6% of revenue in 2024 to 51.5% in the first nine months of 2025, completing the shift in business focus.

 

The prospectus also spells out the risks. Per BigGo Finance's analysis of the latest disclosures, first-half 2026 revenue growth is guided to roughly 35.6% to 45.4% with adjusted net profit expected to decline about 6% to 22%, after first-quarter net profit fell 47.69% year over year. Average humanoid prices dropped from 590,000 to 166,400 yuan, a clear volume-over-price strategy that also compresses future pricing headroom. Unitree acknowledges in the prospectus that it does not fully understand customers' specific use scenarios and that its current advantage rests primarily on cost-effectiveness rather than irreplaceability.

 

 

Beyond the financials, shipment volume offers a more objective comparison. Per Omdia data, the top three humanoid shipment leaders in 2025 were Agibot at 5,168 units, Unitree at 4,200 and UBTech at 1,000, giving UBTech roughly 7.5% market share. It bears noting that methodologies differ across research firms, and Unitree's own prospectus reports more than 5,500 humanoids shipped in 2025 for a world-leading 32.4% share. Under either measure, Unitree leads UBTech by an order of magnitude in units.

 

That gap explains the divergent pricing strategies. Unitree's G1 humanoid retails around $13,500 and the Go2 robot dog starts at $1,600, targeting consumer and developer markets through mass production. UBTech's Walker series serves automotive plants, power grids and similar industrial settings, with high unit prices, long sales cycles and rigorous validation, having won a single order near 100 million yuan in July, delivered over 1,000 Walker units in 2025 and targeting 10,000 units of industrial humanoid capacity in 2026.

 

One sells many units cheaply, the other sells few units expensively. Judging which is stronger is ultimately a judgment about which use case reaches commercial viability first.

 

 

This is a practical problem. UBTech trades in Hong Kong under 9880, accessible through Stock Connect or international brokers. Unitree's subscription opens on 10 August with listing expected in the second half of the month pending exchange confirmation, and the STAR Market imposes thresholds of 500,000 yuan in assets and 24 months of trading experience on mainland investors, while overseas investors cannot subscribe at all.

 

That is the gap stock futures on crypto platforms fill. MEXC lists USDT-settled UNITREE stock futures, and for investors shut out of the A-share subscription, the instrument offers flexibility traditional channels cannot match on several dimensions. On timing, price expectations for Unitree can be traded before the official listing rather than waiting for the debut window. On size, there is no allocation lottery and no subscription cap. On hours, the contract runs 24/7, unconstrained by A-share sessions or market holidays, so overnight and weekend news can be acted on immediately. On direction, both long and short positions are available, and shorting a debut that looks priced on euphoria is effectively impossible with A-share new listings. On capital efficiency, per MEXC's official announcement, maximum leverage on UNITREEUSDT was raised from 10x to 25x on 4 August, letting the same margin control a larger position.

 

It must be stressed that leverage amplifies in both directions. At 25x, a 4% adverse move is enough to wipe out the full margin, and intraday swings in early trading of a hot listing routinely exceed that. MEXC's own announcement warns that leveraged trading involves significant risk and may not suit all investors, and that participants should carefully consider their objectives, experience and risk tolerance beforehand. Leverage magnifies losses as readily as gains, making position sizing more important than directional conviction on instruments like this.

 

 

 

If only one criterion could be used, it should be the sustainability of unit economics. Unitree's 60% gross margin proves hardware alone can make money, but with average prices falling fast, what needs verifying is whether that margin survives the price war. UBTech's 2,203.7% humanoid growth proves industrial demand exists, but what needs verifying is whether scale can carry it across breakeven.

 

On valuation, Unitree's post-issuance base valuation sits near 42 billion yuan, roughly 70 times 2025 adjusted earnings. UBTech's market capitalization has reached around HK$58 billion. Secondary-market expectations for Unitree have been pushed well above the issuance benchmark, with some discussion anchoring beyond 100 billion yuan, though that reflects sentiment rather than official guidance. Against the precedent of ChangXin Memory's 466% first-day surge on 27 July, the scope for sentiment-driven pricing on debut is considerable, and so is the retracement risk.

 

 

 

First, average selling prices and gross margin in Unitree's first post-listing quarterly report. Stabilizing prices would indicate scale economics taking hold, while continued declines would signal an ongoing price war that makes a 60% margin difficult to defend.

 

Second, UBTech's loss narrowing and order conversion. The company recently reported pre-orders exceeding 13,361 units for its U1 consumer humanoid, but the conversion rate into revenue and the trajectory of trade receivables, which management has flagged as a risk, are what determine whether the model works.

 

Third, policy and geopolitical variables. China's directive for 10,000 humanoids deployed in factories, hospitals and disaster response by the end of 2026 provides demand certainty for both. Working the other way, the US has added foreign-made humanoid and quadruped robots to its import restriction list, which weighs more heavily on Unitree given its larger overseas consumer sales.

 

 

For Unitree, the main risks are decelerating growth and valuation detachment, since a debut priced far above fundamentals would face testing every earnings season thereafter, and a 10% free float combined with subscription fever can produce extreme early volatility. For UBTech, the main risks are persistent losses and cash burn, since 3 billion yuan of cumulative losses implies ongoing financing needs, with the stock already under visible pressure this year. For the sector broadly, the shared risk is a commercialization timeline slower than markets expect, as the path from demonstration to volume delivery remains unproven.

 

 

What genuinely matters in this comparison is not whose robot is more advanced but that markets can now test two directly opposing business assumptions against public data. Unitree bets on selling first, trading affordability for shipments and data accumulation while funding itself on hardware margin. UBTech bets on locking in large customers, building a moat through high-value industrial contracts and accepting prolonged losses to secure position. Until now these paths could only be voted on indirectly through private valuations. From August, both get scored continuously against the same public market standard.

 

Two misreadings look likely. The first is equating profitability directly with superiority. Unitree's 600 million yuan profit is genuinely impressive, but it rests on average prices falling from 590,000 to 166,400 yuan, and that profit's durability needs several quarters of verification, while a meaningful share of UBTech's losses reflects capacity building and R&D investment, which differs in character from operating losses. The second is ignoring differences in shipment methodology. The gap between Omdia's 4,200 units for Unitree and the prospectus figure of 5,500 is substantial, and investors citing share data should note the source rather than treating any single measure as settled.

 

What investors should track next is the cross validation of three data sets. Unitree's post-listing average prices and gross margin trajectory, UBTech's order conversion rate and receivables quality, and the change in overseas revenue share at both companies following the US import restrictions. Together these determine whether humanoid robotics is a business taking shape or a narrative still awaiting proof.

 

For cross-asset investors, the broader point is that humanoid robotics completes the AI supply chain from compute to endpoint, and this link is now shifting from private to public pricing. Volatility typically rises before it settles through such a transition, and the transition also brings participation tools outside traditional channels. USDT-settled stock futures let users within a crypto account framework engage these names with more flexibility on timing, direction and capital efficiency, but flexibility has never equaled safety. At 25x leverage, a 4% adverse move erases the margin entirely, and early-session ranges on hot listings routinely exceed that. On instruments like these, defining maximum tolerable loss before sizing a position matters far more than predicting which company wins.

 

 

 

Unitree by a clear margin on profitability. It posted 2025 revenue of 1.708 billion yuan, up 335%, with a 60.27% gross margin and adjusted net profit of 600.1 million yuan, profitable every year since 2020. UBTech generated slightly larger revenue of 2.001 billion yuan, up 53.29%, but with a net loss of 790 million yuan, a 37.7% gross margin and cumulative three-year losses above 3 billion yuan. UBTech leads on revenue scale while Unitree leads decisively on earnings quality.

 

 

Substantially. Per Omdia, the 2025 leaders were Agibot at 5,168 units, Unitree at 4,200 and UBTech at 1,000, leaving UBTech around 7.5% share. Unitree's own prospectus reports more than 5,500 units shipped for a 32.4% world-leading share. Methodologies differ between sources and investors should note which measure they are citing, but Unitree leading UBTech by an order of magnitude in units is not in dispute.

 

 

Pricing and target use cases diverge completely. Unitree pursues affordability, with the G1 humanoid around $13,500 and the Go2 robot dog from $1,600, selling volume into consumer and developer markets as average humanoid prices fell from 590,000 to 166,400 yuan. UBTech targets industrial and commercial deployments, with the Walker series serving automotive plants and power grids at high unit prices and long validation cycles, winning a single order near 100 million yuan in July and targeting 10,000 units of industrial capacity in 2026.

 

 

UBTech trades in Hong Kong under 9880 via Stock Connect or international brokers. Unitree's subscription opens 10 August with listing expected later that month, but the STAR Market requires 500,000 yuan in assets and 24 months of experience from mainland investors, and overseas investors cannot subscribe directly. Derivatives offer an alternative, with MEXC listing USDT-settled UNITREE stock futures that trade before the listing, support both directions and run outside exchange hours, though leveraged products are highly volatile and require full understanding of the risks.

 

 

Flexibility is the main distinction. Price expectations can be traded before the official listing, with no allocation lottery or subscription cap, the contract runs 24/7 regardless of A-share sessions and holidays, and both long and short positions are available, the latter being effectively unavailable on A-share new listings. Per MEXC's official announcement, maximum leverage on UNITREEUSDT rose from 10x to 25x on 4 August, improving capital efficiency. But leverage cuts both ways, and at 25x roughly a 4% adverse move can erase the full margin, so position control is essential.

 

 

The post-issuance base valuation is around 42 billion yuan (about $6.2 billion), roughly 70 times 2025 adjusted earnings. Secondary-market expectations have been pushed well above that benchmark, with some discussion anchoring beyond 100 billion yuan, though this reflects market sentiment rather than official guidance. Against the precedent of ChangXin Memory's 466% first-day surge on the STAR Market on 27 July, the scope for sentiment pricing is significant and so is the retracement risk. Actual trading prices should be treated as authoritative.

 

 

A commercialization timeline slower than markets expect. The path from demonstration to volume delivery remains unproven, and Unitree itself acknowledges in its prospectus that it does not fully understand customers' specific use scenarios and that its advantage derives primarily from cost-effectiveness rather than irreplaceability. The US has also added foreign-made humanoid and quadruped robots to its import restriction list, constraining exports, while China's directive for 10,000 humanoids deployed by end-2026 supports demand. The net effect of these opposing forces remains to be seen.

 

 

This content is provided for informational purposes only and does not constitute investment advice, financial advice, legal advice, tax advice or a recommendation to buy or sell any asset. Prices of equities, new listings, equity derivatives and crypto assets can move sharply, with newly listed stocks and leveraged products particularly volatile and capable of significant gains or losses within very short periods, and leveraged trading can result in the total loss of principal. Past performance, technical indicators and on-chain data cannot guarantee future results, and the financial figures, shipment statistics and valuation discussion presented here come from public sources where methodologies may differ between providers, with listing dates and pricing information subject to the official announcements of the Shanghai Stock Exchange, the Hong Kong Stock Exchange and the relevant companies. Readers should conduct their own research and reach independent conclusions based on their financial circumstances, investment objectives and risk tolerance, consulting licensed professionals where appropriate. The MEXC Crypto Pulse Team accepts no liability for any direct or indirect losses arising from the use of or reliance on this content.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise:

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)