Qfin Holdings, Inc. (NASDAQ:QFIN) Q3 2025 Earnings Call Transcript

Qfin Holdings, Inc. (NASDAQ:QFIN) Q3 2025 Earnings Call Transcript November 18, 2025

Qfin Holdings, Inc. misses on earnings expectations. Reported EPS is $1.52 EPS, expectations were $1.68.

Operator: Ladies and gentlemen, thank you for standing by, and welcome to the Qfin Holdings Third Quarter 2025 Earnings Conference Call. [Operator Instructions] Please also note today’s event is being recorded. At this time, I’d like to turn the conference over to Ms. Karen Ji, Senior Director of Capital Markets. Please go ahead, Karen.

Karen Ji: Thank you, Ken. Hello, everyone, and welcome to Qfin Holdings Third Quarter 2025 Earnings Conference Call. Our earnings release was distributed earlier today and is available on our IR website. Joining me today are Mr. Wu Haisheng, our CEO; Mr. Alex Xu, our CFO; and Mr. Zheng Yan, our CRO. Before we start, I would like to refer you to our safe harbor statement in the earnings press release, which applies to this call as we will make certain forward-looking statements. Also, this call includes discussions of certain non-GAAP financial measures. Please refer to our earnings release, which contains a reconciliation of non-GAAP financial measures to GAAP financial measures. Also, please note that unless otherwise stated, all figures mentioned in this call are in RMB terms. In addition, today’s prepared remarks from our CEO will be delivered in English using an AI-generated voice. Now I will turn the call over to Mr. Wu Haisheng. Please go ahead.

Haisheng Wu: [Interpreted] Hello, everyone. Thank you for joining us today. In the first 9 months of this year, China’s economy and the consumer finance sector have both faced persistent headwinds. The outstanding balance of short-term consumer loans has declined for 3 consecutive quarters on both a year-over-year and quarter-over-quarter basis. Going into Q3, the industry is undergoing a series of regulatory-driven adjustments to improve consumer financial inclusion. We believe these changes will strengthen the sector’s long-term prospects and sustainability, paving the way for healthier and more structured competitive landscape. As such, we view these adjustments not only a challenge but also an opportunity for Qfin. As a leading credit tech platform in China, we continued to prioritize risk management, advance our AI capabilities and deepen collaboration with financial institutions.

We believe these efforts will enable us to better serve inclusive finance needs and strengthen our leadership in the industry. Now I’ll walk you through the progress we made in Q3. By the end of the quarter, our AI-powered credit decision engine and asset distribution platform served 167 financial institutions, delivering efficient, intelligent digital credit services to over 62 million credit line users on a cumulative basis. To navigate the evolving regulatory environment, we dynamically fine-tuned our risk strategies to maintain a healthy balance between risk and growth. As a result, total loan facilitation and origination volume on our platform reached RMB 83.3 billion in the quarter, broadly in line with Q2. Despite the macro headwinds, we delivered steady financial results.

Non-GAAP net income reached RMB 1.51 billion, while non-GAAP EPADS on a fully diluted basis, came in at RMB 11.36, reflecting our solid profitability and operating resilience. On the risk front, funding liquidity in the high-price segment continued to tighten in Q3, leading to an uptick in overall delinquency risk across the industry. To stay closely aligned with evolving market conditions, we further tightened our credit standards and optimized our customer mix by increasing the proportion of high-quality borrowers. In addition, we proactively refined our risk models and completed 611 iterations, implementing differentiated risk management and distribution strategies. On the collection front, we improved efficiency through smarter resource allocation and deeper technology integration.

For example, we allocated more resources to high-performing collection partners to ensure sufficient capacity and better productivity. For customers willing to repay but facing temporary financial difficulties, we offered measured concessions and flexible repayment options. In addition, we were able to assess repayment intent and capacity in real time through large language model algorithms, enabling more precise segmentation and more agile resource deployment. These efforts helped us maintain steady progress even as the broader industry faced rising collection pressure. Our FPD 7, a leading risk indicator for new loans declined in September versus August. Since October, given the new regulations and heightened industry self-discipline initiatives, we expect risk indicators to remain volatile in the near term with current levels above historical averages.

That said, having navigated multiple industry adjustment cycles in the past with prompt and effective responses, we remain confident that we can once again bring risk levels back within a reasonable range in a timely manner. On the funding front, we have been white-listed by all of our active financial institution partners, ensuring a smooth and stable cooperation going forward. Despite a relatively tight funding environment driven by liquidity conditions and policy factors, we maintained the industry-leading pricing power and secured ample funding supply at stable costs. Our average funding cost for Q3 held steady from last quarter, remaining at historical lows. In the ABS market, we issued RMB 4.5 billion during the quarter, up 29% year-over-year with issuance costs down by another 10 basis points.

For the first 9 months of 2025, total ABS issuance grew 41% year-over-year to RMB 18.9 billion, further optimizing our funding structure. Looking ahead, we expect our funding costs to remain largely stable in the coming quarters. For user acquisition, we continue to diversify our channels, enhance targeted operation and improve efficiency compared with last quarter. The number of new credit line users grew by 9% to $1.95 million while average cost per credit line user declined by 8%. The number of new borrowers also grew 10% sequentially to $1.35 million. We have seamlessly integrated convenient and efficient credit services into diversified channels and scenarios, including short-form videos, e-commerce, mobility, food delivery, and financial services.

In Q3, we further expanded our embedded finance network, adding 7 new strategic partners and extending our presence across Internet and financial institution platforms. As a result, the number of new credit line users from the embedded finance channels increased by 13% sequentially, while loan volume up by 11%. For placement strategy, we remain focused on onboarding high-quality users and optimizing our overall user mix. As such, our long-term strategic priority will focus more on our high-quality customers. Supported by AI-driven data models, we expect to gain deeper insights into user needs and behaviors and further refine products and services. This approach will allow us to deliver a superior user experience and improve both our unit economics and user lifetime value.

We believe this focus is critical to strengthening our long-term competitive edge and cementing our leadership position in the industry. In our Technology Solutions business, we continue to advance our AI plus banking strategy, empowering financial institutions in their digital and intelligent transformation. During the quarter, loan volume supported by this business achieved exponential growth, up by roughly 218% on a sequential basis. Our collaboration with banks continue to deepen, expanding from their proprietary channels to a broader range of Internet scenarios where we provide end-to-end technology support in customer acquisition and risk management. Powered by our FocusPRO credit tech platform, our proprietary solution for SME lending, which is built on a 3-tiered credit assessment system, was adopted by several new banking partners and received positive feedback for its industry-leading performance.

As part of our AI plus banking initiative, our 2 proprietary AI agents, the AI Credit Officer and AI Loan Officer, entered pilot testing with our first bank client. The engagement rate among the activated user base has reached around 50%, providing initial validation for the AI agent practical effectiveness in core credit scenarios. Looking ahead, we will focus on strengthening our capabilities in multimodal recognition, voice data collection, lead management and feedback loops while expanding pilot programs and further improving user engagement. At the same time, we are seeing growing interest from financial institutions, laying a strong foundation for broader commercial rollout and scaled adoption in the next phase. On October 1, the new rules officially came into effect.

As a leading player in the industry, we have always held ourselves to the highest compliance standards with no exception this time. Working closely with our financial institution partners, we quickly optimized our business structure and product experience. While these measures may temporarily impact our loan volume and profitability, we believe that prioritizing value for users will eventually strengthen their trust and help us maintain more sustainable and resilient growth over the long term. Meanwhile, certain new industry-wide regulatory measures may have some impact on the industry dynamics. That said, we believe our diversified business model and ample funding capacity will help position us to navigate these changes with limited disruption.

Given the current phase of industry-wide adjustment, we will prioritize risk management over near-term growth, focusing on improving user quality and collection efficiency. Since mid-October, we have already seen encouraging early signs of stabilization in asset quality. Over the years, we have a proven track record of emerging stronger from past challenges, including multiple industry-wide adjustments, and we are confident that this time will be no different. Looking ahead, we will continue to advance our One Body, Two Wings strategy, further strengthen our AI capabilities and empower financial institutions in their digital transformation, driving efficient, healthy and sustainable development of our core business. On the international front, we are actively exploring opportunities across multiple overseas markets.

After extensive research, we are even more convinced that our fintech capabilities are among the best in the world. We view the international expansion as a challenging yet strategically sound path. Quality always comes from deliberate execution, and we are confident we will deliver. In closing, short-term industry headwinds will not alter our long-term trajectory or our fundamental commitment to giving back to our shareholders. Going forward, we will continue to pursue efficient capital allocation and deliver value to our shareholders through compelling shareholder returns. With that, I will now turn the call over to Alex.

Zuoli Xu: Thank you, Haisheng. Good morning, and good evening, everyone. Welcome to our third quarter earnings call. Unexpected China events in the last few months put significant pressure to our operations, and such headwinds may persist through the next couple of quarters as the consumer finance industry faces new round of regulatory scrutiny and the participants try to settle in the vastly different environment. Total net revenue for Q3 was CNY 5.21 billion versus CNY 5.22 billion in Q2 and CNY 4.37 billion a year ago. Revenue from credit-driven service capital heavy was CNY 3.87 billion in Q3 compared to CNY 3.57 billion in Q2 and CNY 2.9 billion a year ago. The sequential and year-on-year increase was mainly driven by higher capital heavy loan balance.

Overall funding costs remained stable Q-on-Q despite some liquidity shortage later in the quarter. In the first 3 quarters, we issued a record-breaking CNY 18.9 billion ABS, an increase of over 40% year-on-year. Revenue from platform service capital light was CNY 1.34 billion in Q3 compared to CNY 1.65 billion in Q2 and CNY 1.47 billion a year ago. The year-on-year and sequential decline was mainly driven by lower capital light facilitation and ICE volume. Platform service account for roughly 48% of our quarter-ending loan balance. We will continue to make timely adjustments to the business mix through the rest of the year to reflect the changing market dynamics and regulatory guidelines. During the quarter, average IRR of the loans we originated and/or facilitated was 20.9% compared to 21.4% in Q2.

Looking forward, we may see further pricing decline as the new regulatory environment requirement being fully implemented across the industry, although the pace of the decline should be modest. Sales and marketing expenses remained stable Q-on-Q, but unit cost declined by about 8% sequentially. We added approximately 1.95 million new credit line users in Q3 versus 1.79 million in Q2. We will likely to adjust the pace of the new user acquisition in the coming months given the volatile macro condition and further optimize our user acquisition channels and improve user engagement and retention. 90-day delinquency rate was 2.09% in Q3 compared to 1.97% in Q2. Day 1 delinquency rate was 5.5% in Q3 versus 5.1% in Q2. 30-day collection rate was 85.7% in Q3 versus 87.3% in Q2.

C-M2, which represents the outstanding delinquency rate after 30 days collection increased Q-on-Q to 0.79% from 0.64%. As overall portfolio risk continued to increase in the last few months, we took additional measures to tighten the risk standard in September and October. While still a bit too early to reverse the trend, we start to see marginal improvement in new loans quality. It may take a few more months to see overall portfolio risk improves as the mix of the loans become more favorable. In such a challenging backdrop, we took even more conservative approach to book provisions against potential credit loss. Total new provisions for risk-bearing loans in Q3 were approximately CNY 2.58 billion versus CNY 2.5 billion in Q2 despite lower risk-bearing loan volume Q-on-Q.

Provision booking ratio hit another historical high. Write-backs of previous provisions were approximately CNY 785 million in Q3 versus CNY 1.18 billion in Q2. Provision coverage ratio, which is defined as total outstanding provisions divided by total outstanding delinquent risk-bearing loan balance between 90 and 180 days, remain near historical high at 613% in Q3. Non-GAAP net profit was CNY 1.51 billion in Q3 compared to CNY 1.85 billion. Non-GAAP net income per fully diluted ADS was RMB 11.36 in Q3 compared to RMB 13.63 in Q2 and RMB 12.35 a year ago. At the end of Q3, total outstanding ADS share count was approximately 130.2 million compared to 132.4 million at the end of Q2 and 144.2 million a year ago. Effective tax rate for Q3 was 20.9% compared to our typical ETR of approximately 15%.

The higher-than-normal ETR was mainly due to withholding tax provision related to the cash distribution from onshore to offshore. With higher contribution from capital heavy model, our leverage ratio, which is defined as a risk-bearing loan balance divided by shareholders’ equity was 3.0x in Q3, still near the low end of historical range. We expect to see leverage ratio fluctuated around this level in the near term. We generate approximately CNY 2.5 billion cash from operations in Q3 compared to CNY 2.62 billion in Q2. Total cash and cash equivalents and short-term investment was CNY 14.35 billion in Q3 compared to CNY 13.34 billion in Q2. Our strong cash flow and financial position should give us sufficient resources to navigate through the challenging environment and allow us to satisfy the commitment and obligations to the market.

We started to execute the $450 million share repurchase program in January 1. As of November 18, 2025, we had in aggregate purchased approximately 7.3 million ADS in the open market for the total amount of approximately CNY 281 million, inclusive of commissions at the average price of USD 38.7 per ADS. We intend to resume the repurchase program after the window opened after this earnings call. Finally, regarding our business outlook. Given the persistent economic uncertainty and fast-changing market dynamic, we will continue to take a cautious approach in business planning for the next couple of quarters, focusing on risk control of our operation. For the fourth quarter of 2025, the company expects to generate non-GAAP net income between CNY 1 billion and CNY 1.2 billion.

This outlook reflects the company’s current and preliminary view, which is subject to material changes. With that, I would like to conclude our prepared remarks. Operator, we can now take some questions.

Q&A Session

Operator: [Operator Instructions] For those who can speak Chinese, please start your question in Chinese, followed by English translation. [Operator Instructions] Your first question today comes from Chiyao Huang from Morgan Stanley.

Chiyao Huang: [Foreign Language] So basically 2 questions from me. One is after the new loan facilitation come into effect in October, how should the management think about the change to the business model or profit model of the loans? And what’s the expectation for the take rate in 2026? And maybe over the long run, how should we think about the loan economics when they normalize? And number two is how do management think about the competitive landscape after the loan facilitation rule taking effect?

Zuoli Xu: Thank you, Zheong. And in terms of regulation and take rate, with the new rules in place, both on loan facilitation space and the broader consumer finance industry will need some time to adjust. In near term, the rules will have some impact on market size, risk levels and profitability. This is for sure. But in the long run, we believe the competitive environment will become more sustainable and healthier, which is good to our industry. As for the near-term impact, let me talk about what we are seeing right now. First, as the entire industry is lifting the risk bar, funding capacity for our ICE and referral businesses will come down. This means some users will no longer be served, and this will have some impact on our loan volume.

For the rest of ICE business, as we adjust pricing, the take rates will decline. Also on the positive side, we expect to see better conversion, higher loan amounts and less early repayment. This will help you reduce some of the pressure on the net take rate. Second, the liquidity pressure in the market is pushing overall risk higher for the broader consumer finance space. Our C2M2 was up to 0.79% in Q3 from 0.64% in Q2, and the net provisions were up about 36% compared to Q2. We expect this trend to continue at least in the next 1 or 2 quarters. Based on our Q4 guidance, we are roughly talking about take rate of 3% to 4% because of pricing and the risk impact. Over the next 2 quarters, we expect the industry to remain volatile, and we are trying to get a better understanding on our take rate for in the new loan.

For 2026 and beyond, the take rates will depend on how things evolve from the Q4 baseline. Specifically, our focus will be a few things. First, we will continue to optimize our risk strategies and improve collection efficiency to enhance our risk performance. Second, we will further optimize costs in user acquisition and operations to improve overall efficiency. Third, we will also explore some new service offerings to further improve user conversion and retention. We hope these efforts could help improve our take rate over time. And for your second question, for the competitive landscape, since the new rules came out in April, we have seen a major shakeup in the high pricing segment. New loan volumes in that market decreased a lot. Some smaller platform may not survive in the future.

The rest of the platform are also shrinking their loan book. So entering Q4, we are actually seeing less competition for traffic. Looking ahead, some of the platform currently operating in high pricing segment may also try to move into the 18% to 24% range, but it is very difficult for them to be profitable in that band, given their disadvantage in funding risk management and operational efficiency. So in longer term, we think some of these players will eventually leave the market. We think that the market consolidation will benefit us in a few ways. With fewer smaller platforms competing for traffic, our marketing efforts will be more effective. We can acquire higher-value users more accurately with lower acquisition costs. In the new market environment, the user’s multi-borrowing situation improves.

We should be able to expect lower credit risk and better conversion rates. As such, users’ lifetime value will improve in the longer term. So overall, we think the longer-term competitive environment will become more in our favor, and we see room to take more market shares over time. Thank you.

Operator: Your next question comes from Lincoln Yu at JPMorgan.

Lihan Yu: [Foreign Language] I will translate my question. So my question is on shareholder return. So given the recent share price volatility and the regulatory uncertainties, would there be any change in the company’s execution of the existing buyback plans? As I see, we still have about like 170 million remaining from the plan announced like in last November. And also in longer term, what is the company’s consideration on shareholder return?

Zuoli Xu: Okay. Lincoln, I will take this question then. So just like you said, as of now, we still have about 170 million left under our 450 million program designed for this year. And we took a temporary pause during the third quarter, just given the incoming regulatory update and all the risk associated with that. Now after today’s earnings call, the new window will open in terms of repurchase. We will resume the execution of this program to fulfill our commitment for the rest of the year. And then regarding the dividend, we have been stated that our goal is to gradually increase dividend per ADS through the — through each semiannual kind of a dividend payout. And right now, the Board-approved dividend payout ratio is 20% to 30%, which still gives us enough room to maintain that kind of a progressive dividend trend, even with the volatile kind of earnings movement for the next few quarters there.

Eventually, we still aim to achieve that progressive dividend target for the foreseeable future. In the long run, we still put the shareholder return as one of the top priorities for this company, although the mix between the buyback and dividend payout may change from time to time depending on the situation that we are facing at any given time. Thank you.

Operator: Your next question comes from Alex Ye at UBS.

Xiaoxiong Ye: [Foreign Language] So my question is regarding the asset quality trend. So just wondering how has been the trend — monthly trends for October and September and November? Have we seen any rate deterioration versus Q3? And assuming there’s no further trends in regulatory framework, so how — when does management expect the equity to stabilize and pick? What are the upside that we should be aware of?

Haisheng Wu: [Foreign Language]

Karen Ji: [Interpreted] So let me do the translation. Since the new rules started to take effect on October 1, high-cost fundings have tightened further. At the same time, industry risk levels have been going up in Q3. So pretty much all platforms, no matter the price level, have made risk management first and tightened their risk policies. This has made liquidity even tighter and pushed over risk levels further up. But we are also seeing some positive signs in November. The early risk indicators of new loans are showing signs of stabilization and slight improvement. The FPD7 delinquency rate for new loans in September decreased by 8% compared to that of July. In terms of the risk performance of overall loan portfolio, the 7-day delinquency rate observed in November has remained broadly flat compared to October with no further upward trend.

Haisheng Wu: [Foreign Language]

Karen Ji: [Interpreted] So right now, we mainly focus on 2 areas to lower rates. For pre and in loan processes, we are modestly increasing the share of high-quality users to optimize overall rate structure. We are also increasing operational resources for low-risk users and use large language model algorithms to improve pricing. With more tailored pricing, exclusive benefits and a simpler user journey, we intend to improve user conversion and retention. For collection, we are adding more in-house capacity and increasing support for our partner agencies. We are also improving how we profile users and match cases. So each case can go to direct team. Powered by large language algorithms, we can now get a better read on borrowers’ ability and willingness to repay, addressing their grouping and tailor our approach to drive better [indiscernible].

Haisheng Wu: [Foreign Language]

Karen Ji: [Interpreted] So looking ahead, although we have seen some early signs of stabilization, it’s only been about 2 weeks into November. So we will need some more time to tell if the trend will hold. Our loan tenure is usually 9 to 10 months. So when we tighten risk strategies for new loans, it usually takes 2 to 3 quarters for the improvement to show up in the overall portfolio. But the market dynamic is still evolving, and the leading risk indicators for new loans haven’t been down to our desired levels yet, so this adjustment cycle will likely take a bit longer than we expected. On the financial side, our provisions and profit buffer of our business are both very solid. This gives us plenty of room to manage through the short-term industry headwinds. We have been through many challenges before. At each time, we were able to respond quickly and effectively. So we are confident we can bring risk levels back to a reasonable range once again.

Operator: Your next question comes from Emma Xu of BofA Securities.

Emma Xu: [Foreign Language] So according to recent media reports, regulators are starting new regulations for consumer finance companies that will lower the APR of newly issued loans to 20%. So although these regulations will not apply to loan facilitation firms, has the management evaluated the potential implications if the average APR will fall to below 20%? Could this lead to a slowdown in loan growth and an increase in credit cost? In such a scenario, does the company has any measures in place to hedge against the impact on profitability?

Zuoli Xu: Emma, let me take this one. Yes, on the pricing guidance for consumer finance companies, there’s no formal document [ tariff ] at this point, just informal communication. As we understand, consumer finance companies are required to keep their average pricing below 20%. We think the logic behind this is quite close to the new rules on loan facilitation sector as the regulators’ intention is also to reduce the borrowing costs for consumers and make credit more accessible. In the near term, yes, it will have some impact on market size, risk levels and profitability. But over time, we think it will help create healthier competition and improve asset quality. In terms of funding, our direct exposure to consumer finance companies is small.

So the direct impact on us is limited. First, the consumer finance companies source their business from diverse channels. Industry-wide, about 40% of their loans is self-operated and about 60% from API channels, mostly platform under other Internet companies. Our cooperation with them just accounts for a very small part. In terms of funding, they only account for about 15% of our loan mix. Most of our funding comes from banks. So we are flexible to shift our funding structure if needed. As such, we think the direct impact on us is quite limited, but there is indirect impact. As consumer finance companies adjust their pricing, we may expect further pressure on liquidity in the short term, leading to risk volatility. In that case, we may continue to lift our bar to mitigate the risk.

Our average APR in Q3 was 20.9%. Going forward, we need to strengthen our ability to serve higher-quality users. With a broader user base and a better mix, we should be able to optimize pricing and keep our risk well balanced. In the meantime, we will maintain our operation to improve overall profitability. The point is we care about — we care more about our users’ long-term value than certain profitability. Thank you.

Operator: Your next question comes from Cindy Wang at China Renaissance.

Yun-Yin Wang: [Foreign Language] I have 2 questions here. First, during the opening remarks, CEO mentioned Technology Solutions loan volume up more than 200% quarter-over-quarter in Q3. What’s the main drivers behind it? And what is the outlook of this business? Second, in Q3, capital light accounted for 42% of the new loan volume, largely the same as Q2, but down 3 percentage points quarter-over-quarter to 48% of loan balance. So how do you expect the ratio of capital-heavy and capital-light business to new loan volume and loan balance in Q4 and 2026?

Haisheng Wu: Thank you, Cindy. I can take a first one, and Alex, you can take the second one. So far — yes, so far, our Technology Solutions business has partnered with over 20 financial institutions. In Q3, we facilitated around RMB 5.4 billion in loan volume through this model, up 218% quarter-on-quarter. And the outstanding balance has exceeded RMB 10 billion lately. Two main factors are driving this growth. First, loan volume with our same partners is steadily ramping up. Second, we are expanding the way we collaborate with financial institutions. Not only can we facilitate credit business within their ecosystem, but also across a broader set of online scenarios. This really highlights the value we bring in customer acquisition and risk management across diverse channels.

We are also seeing strong demand from financial institutions for AI agents. Because of that, our solution is more than technology infrastructure. We are currently upgrading our FocusPRO product into our super credit AI agent. Take our AI Credit Officer as an example. Traditional off-line credit products in banks have long complicated processes. Powered by large language model capabilities, AI Credit Officer can use the one single model to handle all kinds of documents processing tasks during due diligence and credit approval states. This will streamline the process by removing overlapping models running in parallel. As a result, users do not need to resubmit their materials. The whole process can be accelerated and the approvals can be completed within the same day.

On the risk assessment side, by leveraging our trillion-level risk decision data sets and multi-model large language model technology, the agent can identify risk in seconds, generate more precise user profiles within minutes and keep iterating based on feedback. In the pilot run with our bank partners, our AI agents are already making an impact in key areas like customer acquisition and approvals. The market feedback has also been very positive. We are also seeing interest from several other financial institutions in their products. We believe the future upside of our super credit AI agent is very huge. Thank you.

Zuoli Xu: Cindy, to your second question regarding the mix between capital heavy and capital light. In the short term, as we are facing very volatile kind of market condition that we discussed earlier, we may need to make some flexible adjustments to the mix. On one hand, for example, in this kind of generally higher risk environment, we intend to do more capital light versus capital heavy. But on the other hand, the price cap on the ’24 also limited our capability to do the ICE side of the business. So those 2 forces probably will work together in the fourth quarter in particular. But directionally, I would say you probably will see a little bit more on the capital light side in the fourth quarter and — as we intend to reduce the risk exposure.

And then the longer term, I think we still need to make from quarter-to-quarter or time to time, we still need to make timely adjustments based on the conditions we were facing based on the risk level the market presents and also based on the funding sources we’re getting to decide what’s the best solution or best mix for us in terms of mix. So I don’t think there will be — at least for the 2026, I don’t think there will be a directional movement toward the light or towards the heavy, and most likely, we’ll be sort of bouncing around the sort of the 50-50 line throughout the next year. Thank you.

Operator: Thank you. That concludes our question-and-answer session for today. I’d like to hand back for closing remarks. Thank you.

Zuoli Xu: Okay. Thank you again for everyone to join us for the call. If you have additional questions, please feel free to contact us offline. Thank you. Have a good day.

Operator: Thank you. That does conclude our call for today. You may now disconnect your lines. Thank you. [Portions of this transcript that are marked [Interpreted] were spoken by an interpreter present on the live call.]