Q2 2026 Tencent Holdings Ltd Earnings Call

August 12, 2026

Speaker #1: Just know her song, she's so toward, she means something in young. It's only for the true love.

Wendy Huang: Good day and good evening. Thank you for standing by. Welcome to Tencent Holdings Limited 2026 second quarter results announcement webinar. I am Wendy Huang from Tencent IR team. At this time, all participants are in a listen-only mode. After management's presentation, there will be a question and answer session. For participants who dial in by phone, if you wish to ask a question, please press five on your telephone to raise your hand. If you are accessing from the Tencent Meeting or Room Meeting application, please click the raise hand button at the bottom left. Please be advised that today's webinar is being recorded. Before we start the presentation, we would like to remind you that it includes forward-looking statements, which are underlined by a number of risks and uncertainties and may not be realized in the future for various reasons.

Speaker #2: Good evening, and thank you for standing by. Welcome to the Tencent Holdings Limited Q2 2026 results announcement webinar. I'm Wendy Huang from the Tencent IR team.

Speaker #2: At this time, all participants are in a listen-only mode. After management's presentation, there will be a question-and-answer session. For participants who dialed in by phone, if you wish to ask a question, please press 5 on your telephone to raise your hand.

Speaker #2: If you are accessing from the Tencent meeting or room meeting application, please click the raise hand button at the bottom left. And please be advised that today's webinar is being recorded.

Speaker #2: Before we start the presentation, we would like to remind you that it includes forward-looking statements, which are underlined by a number of risks and uncertainties, and may not be realized in the future for various reasons.

Speaker #2: Information about general market conditions is coming from a variety of sources outside of Tencent. This presentation also contains some unaudited non-IFRS financial measures that should be considered in addition to, but not as a substitute for, measures of the group's financial performance reported in accordance with IFRS.

Wendy Huang: Information about general market conditions is coming from a variety of sources outside of Tencent. This presentation also contains some unaudited non-IFRS financial measures that should be considered in addition to, but not as a substitute for, measures of the group's financial performance prepared in accordance with IFRS. For a detailed discussion of risk factors and non-IFRS measures, please refer to our disclosure documents on the IR section of our website. Let me now introduce the management team on the webinar tonight. Our Chairman and CEO, Pony Ma, will kick off with a short overview. President Martin Lau will provide a strategy review. Chief Strategy Officer James Mitchell will provide a business review, and Chief Financial Officer John Lo will conclude with financial discussion before we open the floor for questions. I will now pass it to Pony.

Speaker #2: For a detailed discussion of risk factors and non-IFRS measures, please refer to our disclosure documents on the IR section of our website. Let me now introduce the management team on the webinar tonight.

Speaker #2: Our chairman and CEO, Pony Ma, will kick off with a short overview. President Martin Lau will provide a strategy review. Chief Strategy Officer, James Mitchell, will provide a business review.

Speaker #2: And Chief Financial Officer, John Lau, will conclude with financial discussion before we open the floor for questions. I will now pass it to Pony.

Speaker #3: Thank you, Wendy. Good evening. Thank you, everyone, for joining us. As we enter the third quarter of the year, we are making substantial progress toward building a new AI-empowered Tencent in terms of intelligence, applications, and infrastructure.

Pony Ma: Thank you, Wendy. Good evening. Thank you everyone for joining us. As we enter into the Q3 of the year, we are making substantial progress toward building a new AI-empowered Tencent in terms of intelligence, applications, and infrastructure. At the intelligence level, Hunyuan 3.0 production version provide user a wide use model with a strong cost performance metrics, and serve as a stepping stone toward the Hunyuan family of models, attaining state-of-the-art capabilities in the future. At the application level, our WorkBuddy AI office, a productivity service, and CodeBuddy AI coding tool are achieving breakout user growth and are the clear leaders in the field in China today. At the infrastructure level, we substantially expanded our procurement of compute, which will enable us to convert usage of our applications and models into revenue going forward.

Speaker #3: At the intelligence level, Wen Yuan's production version provides users a wide-use model with strong cost-performance metrics, and is steadily stepping forward as part of the Wen Yuan family of models, attending state-of-the-art capabilities in the future.

Speaker #3: At the application level, our body AI office productivity service and core body AI coding tool are achieving breakout user growth and are the clear leaders in the field in China today.

Speaker #3: At the infrastructure level, we substantially stepped up our procurement of compute, which will enable us to convert usage of our applications and models into revenue going forward.

Speaker #3: At the same time, we continue to enhance our existing services, with substantial marketing service revenue growth, several successful recently released games, and rapidly increasing video views on our video accounts.

Pony Ma: At the same time, we continue to enhance our existing services with a sustained marketing service revenue growth, several successful recently released games, and rapidly increasing video views on Weixin Video Accounts. Looking at our financial numbers for the Q2. Total revenue was RMB 205 billion, up 11% year-on-year. Gross profit was RMB 118 billion, up 13% year-on-year. Non-IFRS operating profit was RMB 76 billion, up 9% year-on-year. Excluding new AI products, non-IFRS operating profit was RMB 86 billion, up 19% year-on-year. Non-IFRS net profit attributable to equity holders was RMB 68 billion, up 9% year-on-year. Turning to our key services. For communication and social networks, combined MAU of Weixin and WeChat grew year-on-year and quarter-on-quarter to 1.4 billion. For digital content, TME's acquisition of Ximalaya strengthened our audio content and unlocked new synergies with the ecosystem IPs, including from China Literature.

Speaker #3: Looking at our financial numbers for the second quarter, total revenue was 205 billion RMB, up 11% year on year. Gross profit was 118 billion RMB, up 13% year on year.

Speaker #3: Non-IFRS operating profit was RMB 76 billion, up 9% year-on-year. Excluding new AI products, non-IFRS operating profit was RMB 86 billion, up 19% year-on-year.

Speaker #3: And non-IFRS net profit attributed to equity holders was 68 billion RMB, up 9% year on year. Turning to our key services, for computation and social networks, combined MAU of reaching a WeChat group year on year and quarter on quarter to 1.4 billion.

Speaker #3: For digital content, TMU's acquisition of Himalaya Sensing, our audio content and unlocked new signatures with the ecosystem IPs, including from China literature. For games, new game local kingdom world ranked first by average DAU and by gross receipt.

Pony Ma: For games, new game, Roco Kingdom: World was ranked first by average DAU and by gross receipts among all new games released in China industry-wide this year. For cloud, WorkBuddy recently ranked first among productivity AI service in China based on monthly interactions. I will now hand it over to Martin for the strategy review.

Speaker #3: Among all new games released in China industry-wide this year, for cloud, WorkBody recently ranked first among productivity AI services in China based on monthly interactions. I will now hand over to Martin for the strategy review.

Speaker #3: Thank you, Pony, and good evening and good morning to everybody. Today, we want to provide you with an update on our overall AI strategy.

Martin Lau: Thank you, Pony, and good evening and good morning to everybody. Today, we want to provide you with an update on our overall AI strategy. Tencent's existing businesses are growing solidly due to intrinsic modes and AI enablement. As discussed earlier this year, our modes arise from factors including network effect, depth, and value added along supply chain, IP, low take rates, regulatory requirements, and private data. In addition to these modes, we are further deploying AI to boost returns in areas including Weixin, games, and advertising. As a result, our existing businesses provide a very strong financial support for our new AI initiatives.

Speaker #3: Tencent's existing businesses are growing solidly due to intrinsic moats and AI enablement. As discussed earlier this year, our moats arise from factors including network effect, depth, and value added along the supply chain.

Speaker #3: IP, low tick rates, regulatory requirements, and private data. In addition to these modes, we're further deploying AI to boost returns in areas including wage and games, and advertising.

Speaker #3: As a result, our existing businesses provide a very strong financial support for our new AI initiatives. Regarding our new AI initiatives, we've made significant progress in constructing a robust foundation.

Martin Lau: Regarding our new AI initiatives, we have made significant progress in constructing a robust foundation, including a substantially improved Hunyuan 3 foundation model with leading cost performance, WorkBuddy and CodeBuddy that lead the China market in terms of AI productivity usage, and Yuanbao and Xiaowei serving as gateways to drive broader consumer AI adoption. We see increasing potential to generate attractive financial returns from franchise products with differentiated advantages, including Hunyuan, WorkBuddy, and Xiaowei over time. We are comfortable in making significant investments in AI because not only there is a substantial upside potential, there is also clear downside protection. The AI investments we are making are mostly in AI infrastructure, and in the worst case, which we do not believe that would happen, we can choose to rent that infrastructure out at cost recovery or even better prices via Tencent Cloud if needed. Now, going on to the different components.

Speaker #3: Including a substantially improved Wen Yuan 3 Foundation model with leading cost performance, work body and core body that lead the China market in terms of AI productivity usage, and Yuanbao and Xiaowei serving as gateways to drive broader consumer AI adoption.

Speaker #3: We see increasing potential to generate attractive financial returns from franchise products with differentiated advantages, including Wen Yuan work body and Xiaowei over time. We're comfortable in making significant investments in AI because not only there is a substantial upside potential, there's also clear downside protection.

Speaker #3: The AI investments we're making are mostly in AI infrastructure and in the worst case, which we do not believe that would happen, we can choose to rent that infrastructure out at cost recovery or even better prices via Tencent Cloud if needed.

Speaker #3: Moving on to the different components, first, on the Wen Yuan Foundation model: The release of Wen Yuan 3's full production version has been very successful, showing a substantial step up in performance compared to the Wen Yuan 3 preview version.

Martin Lau: First, on Hunyuan foundation model. The release of Hunyuan 3's full production version is very successful, showing a substantial step-up in performance compared to the Hunyuan 3 preview version. Leveraging the feedback loop from product teams to improve the quality and diversity of data used for post-training, and by scaling up reinforcement learning, Hunyuan 3 achieved a notable improvement in task completion rates and meaningful reduction in hallucination and error rates. The improvement in Hunyuan 3's capabilities are most evident in its agentic capabilities and product experience. The model's performance step-up across reasoning, agentic, and long context tasks delivered clear advantages for use cases such as coding, office work, financial modeling, and front-end design.

Speaker #3: Leveraging the feedback loop from product teams to improve the quality and diversity of data used for post-training, and by scaling up reinforcement learning, Wen Yuan 3 achieved a notable improvement in task completion rates and a meaningful reduction in hallucination and error rates.

Speaker #3: The improvements in Wen Yuan 3's capabilities are most evident in its AGN capabilities and product experience. The model's performance step-up across reasoning, AGN, and long context tasks delivered clear advantages for use cases such as coding, office work, financial modeling, and front-end design.

Speaker #3: These performance improvements drove accelerated user adoption and growing external customer demand, validating its practical utility in real-world usage, as demonstrated by the approximately six-times increase in average daily tokens usage of Wen Yuan 3 compared to the preview version across all channels during the pay period.

Martin Lau: These performance improvements drove accelerated user adoption and growing external customer demand, validating its practical utility in real-world usage, as demonstrated by the approximately 6 times increase in average daily tokens usage of Hunyuan 3 compared to the preview version across all channels during the pay period. Additionally, Hunyuan 3 consistently ranks among the top three models globally on OpenRouter based on token usage. Hunyuan 3's production version has performed well and will serve as a stepping stone toward the Hunyuan family of models achieving state-of-the-art capabilities in the future. While providing users with the cost performance efficiency that they need today. We have been integrating Hunyuan into our products, making great impact. For WorkBuddy, Hunyuan can facilitate complex agent workflows with higher task success rates and reduced time to completion. For Yuanbao, Hunyuan delivers leading execution quality in information retrieval, data processing, document workflows, and everyday decision-making.

Speaker #3: Additionally, Wen Yuan 3 consistently ranks among the top three models globally on open router based on token usage. Wen Yuan 3's production version has performed well and will serve as a stepping stone toward the Wen Yuan family of models achieving state-of-the-art capabilities in the future.

Speaker #3: While providing users with the cost performance efficiency that they need today, we have been integrating Wen Yuan into our products, making great impact. For the work body, Wen Yuan can facilitate complex agent workflows with higher task success rates and reduced time to completion.

Speaker #3: For Yuanbao, Wen Yuan delivers leading execution quality in information retrieval, data processing, document workflows, and everyday decision-making. In games, we're leveraging Wen Yuan for AI teammate creation and code review for games, including our flagship game, Peacekeeper Elite.

Martin Lau: In games, we are leveraging Hunyuan for AI teammate creation and code review for games, including our flagship game, Peacekeeper Elite. In Weixin, we deployed Hunyuan for powering the AI assistant in official accounts and the developer tools for Mini Programs. At the same time, product integration is making Hunyuan better by continuously feeding real-world product usage and domain feedback into model training. Our model product co-design approach allows Hunyuan to validate model accuracy and identify and work on edge cases, enabling faster model iteration and sustained performance gains. Having now established a new system for fast model iteration, and with the Hunyuan 3 validating the system, we are accelerating the improvement of our model. We are scaling more powerful reinforcement learning to substantially upgrade models after pre-training is done, and we are in the process of upgrading multimodal capabilities.

Speaker #3: And in Weixin, we deployed Wen Yuan for powering the AI assistant in official accounts and the developer tools for mini programs. At the same time, product integration is making Wen Yuan better.

Speaker #3: By continuously feeding real-world product usage and domain feedback into model training, our model product co-design approach allows Wen Yuan to validate model accuracy and identify and work on edge cases.

Speaker #3: Enabling faster model iteration and sustained performance gains. Having now established a new system for fast model iteration, and with Wen Yuan 3 validating the system, we're accelerating the improvement of our model.

Speaker #3: We're scaling more powerful reinforcement learning to substantially upgrade models after pre-training is done. And we're in the process of upgrading multimodal capabilities. More importantly, we're training a larger parameter model Wen Yuan 4, which we expect to release later this year.

Martin Lau: More importantly, we are training a larger parameter model, Hunyuan 4, which we expect to release later this year. By accelerating the technical iteration and pushing the boundaries of model intelligence, we are confident Hunyuan's capabilities will reach state-of-the-art level. We believe we will generate significant return in building a large and valuable AI-native new business for Tencent. The rationale behind investing in our own foundation model is that we can achieve better unit economics, more innovative features, and more exposure to the value of intelligence through co-design across our applications, our model, and our compute infrastructure, especially at this early stage of AI diffusion. Moving on to the application front. Our AI office productivity workspace, WorkBuddy, and coding tool, CodeBuddy, are achieving breakout success in terms of capability and user growth. They are the clear leading office productivity service in China based on monthly interactions.

Speaker #3: By accelerating the technical iteration and pushing the boundaries of model intelligence, we're confident Wen Yuan's capabilities will reach a state-of-the-art level. We believe we will generate a significant return in building a large and valuable AI-native new business for Tencent.

Speaker #3: The rationale behind investing in our own foundation model is that we can achieve better unit economics, more innovative features, and more exposure to the value of intelligence through co-design across our applications, our model, and our compute infrastructure.

Speaker #3: Especially at this early stage of AI diffusion. Moving on to the application front, our AI office productivity workspace, WorkBody, and coding tool, CoreBody, are achieving breakout success in terms of capability and user growth. They are the clear leading office productivity services in China based on monthly interactions.

Speaker #3: Work body serves as a one-stop shop workspace that orchestrates multiple agents to handle complex work from end to end. Users can remotely control work body via waging and Wecom as well as VPC and access to over 70,000 skills from Tencent Cloud skill hub.

Martin Lau: WorkBuddy serves as a one-stop shop workspace that orchestrates multiple agents to handle complex work from end to end. Users can remotely control WorkBuddy via Weixin and WeCom, as well as the PC, and access to over 70,000 skills from Tencent Cloud SkillHub. Besides the rapid user adoption of WorkBuddy, it is also achieving high retention rates and high willingness to pay among users as it directly contributes to users' productivity. It also attracts growing and more vibrant developer community by embedding skill pay and Weixin Pay inside task flows to enable payouts for developers when their skills are called. This progress supports our view that there are substantial opportunities to be unlocked in the productivity market, including coding and existing office work scenarios. We're currently focused on investing in market education and extending our market leadership position.

Speaker #3: Besides the rapid user adoption of Workbody, it is also achieving high retention rates and a high willingness to pay among users, as it directly contributes to users' productivity.

Speaker #3: It also attracts a growing and more vibrant developer community by embedding skill pay and wage pay inside task flows to enable payouts for developers when their skills are called.

Speaker #3: This progress supports our view that there are substantial opportunities to be unlocked in the productivity market, including coding and existing office work scenarios. We're currently focused on investing in market education and extending our market leadership position.

Speaker #3: Over time, product economics will be attractive, as enhanced premium benefits accelerate paying user growth, while we can reduce token costs through agent efficiency, inference efficiency, and model optimization.

Martin Lau: Over time, product economics will be attractive as enhanced premium benefits accelerate paying user growth while we can reduce token costs through agent efficiency, inference efficiency, and model optimization. Given Tencent applications such as Weixin, WeCom, and Tencent Meeting are already widely used by enterprises, WorkBuddy provides a new way for us to monetize our enterprise relationships. On the consumer front, we recently released a prototype of Xiaowei, which delivers an embedded and context-aware agentic AI experience within Weixin, leveraging Weixin's social graph, knowledge graph, merchant reach, and payment functionality. Xiaowei is powered by the Weixin customized model, WeLM, built with a focus on user privacy, Weixin-specific use cases, and cost efficiency. Xiaowei can help users navigate and derive insights from Weixin's diverse content universe in a personalized and efficient manner.

Speaker #3: Given Tencent applications such as WeGame, WeCom, and Tencent Meeting are already widely used by enterprises, Workbody provides a new way for us to monetize our enterprise relationships.

Speaker #3: On the consumer front, we recently released a prototype of Xiaowei, which delivers an embedded and context-aware AGN AI experience within Weixin, leveraging Weixin's social graph, knowledge graph, merchant reach, and payment functionality.

Speaker #3: Xiaowei is powered by the Weixin customized model WLM, built with a focus on user privacy, Weixin-specific use cases, and cost efficiency. Xiaowei can help users navigate and derive insights from Weixin's diverse content universe in a personalized and efficient manner. Xiaowei can also leverage Weixin's unique mini program ecosystem to help users discover products, make purchase decisions, and place orders, laying the groundwork for an agent-to-agent transaction model.

Martin Lau: Xiaowei can also leverage Weixin's unique Mini Program ecosystem to help users discover products, make purchase decisions, and place orders, laying the groundwork for an agent-to-agent transaction loop. While the prototype can technically handle advanced agentic workflows, we currently configure Xiaowei to require user intervention and multiple step confirmations as safety measures. Xiaowei will be rolled out to broader user base in a phased approach as we work on several core initiatives to elevate the user experience. These include upgrading Xiaowei's dialogue, memory, and recommendation capabilities, expanding service and content integrations, scaling our AI infrastructure, and upgrading our harness to support a significantly larger user base.

Speaker #3: While the prototype can technically handle advanced AGN workflows, we currently configure Xiaowei to require user intervention and multiple step confirmations as safety measures. Xiaowei will be rolled out to broader user base in a phased approach as we work on several core initiatives to elevate the user experience these include upgrading Xiaowei's dialogue memory and recommendation capabilities, expanding service and content integrations, scaling our AI infrastructure, and upgrading our harness to support a significantly larger user base.

Speaker #3: As we upgrade waging for the AI era, we can do it in a cost efficient way and we're confident that AI will over time accelerate the growth and thus the monetization of the entire waging ecosystem, generating attractive return for us.

Martin Lau: As we upgrade Weixin for the AI era, we can do it in a cost-efficient way, and we're confident that AI will, over time, accelerate the growth and thus the monetization of the entire Weixin ecosystem, generating attractive return for us. Turning to Yuanbao, we are focusing on improving its capabilities and user experience, particularly in search, speech recognition, and text-to-speech functionality. We are also improving its ability to address broader long-tail AI needs of consumers, including multimodal generation. Yuanbao plays an important role in the co-design flywheel as its conversational use cases generate valuable feedback to help improve our Hunyuan family of models. Over time, functionalities developed and honed by Yuanbao can become atomic capabilities for use in other Tencent products such as WorkBuddy, CodeBuddy, Weixin, and QQ Browser. With that, I pass on to James.

Speaker #3: Turning to Yuanbao, we are focusing on improving its capabilities and user experience, particularly in search, speech recognition, and text-to-speech functionality. We're also enhancing its ability to address the broader, long-tail AI needs of consumers, including multimodal generation.

Speaker #3: Yuanbao plays an important role in the co-design flywheel as its conversational use cases generate valuable feedback to help improve our Wen Yuan family of models.

Speaker #3: Over time, functionalities developed and honed by Yuanbao can become atomic capabilities for use in other Tencent products such as work buddy, code buddy, waging, and QQ browser.

Speaker #3: And with that, I pass on to James.

Speaker #1: Thank you, Martin, for the quarter total revenue was up 11% with social networks contributing 16%, domestic games 23%, international games 9%, marketing services 21%, and fintech and businesses services 30%.

James Mitchell: Thank you, Martin. For the quarter, total revenue was up 11%, with Social Networks contributing 16%, Domestic Games 23%, International Games 9%, Marketing Services 21%, and FinTech and Business Services 30%. Our gross profit was up 13%, within which VAS gross profit increased 14%, Marketing Services 21%, and FinTech and Business Services 9%. Value-added service revenue was 98 billion RMB, up 8% year-on-year. Within which the Social Network revenue was up 1% to 32 billion RMB, driven by increased revenue from app-based game item sales, partially offset by decreased revenue from long-form video subscriptions, where revenue decreased 6% year-on-year. However, our exclusive drama series, "The Lead," was the most-watched drama series across all video platforms in China in Q2. Audio subscription revenue increased 8%, driven by higher music ARPU and enriched content stemming from the inclusion of Ximalaya.

Speaker #1: Our gross profit was up 13%. Within that, VAS gross profit increased by 14%, marketing services by 21%, and fintech and business services by 9%. Value-added services revenue was RMB 98 billion, up 8% year on year.

Speaker #1: Within which the social network revenue was up 1% to 32 billion renminbi, driven by increased revenue from app-based game item sales, partially offset by decreased revenue from long-form video subscriptions, where revenue decreased 6% year on year.

Speaker #1: However, our exclusive drama series, The Lead, was the most-watched drama series across all video platforms in China in the second quarter. Audio subscription revenue increased 8%, driven by higher music RPU and enriched content stemming from the inclusion of Shimalaya.

Speaker #1: We facilitated users discovering new music by enabling one-click access from video accounts to QQ Music. In May, we completed the acquisition of Shimalaya by bringing Shimalaya into the Tencent Group.

James Mitchell: We facilitated users discovering new music by enabling one-click access from Video Accounts to QQ Music. In May, we completed the acquisition of Ximalaya. By bringing Ximalaya into Tencent Group, we can enhance Tencent Music's resilience, deepen the content supply relationship between China Literature and Ximalaya, and provide users with new content formats, including audiobooks and podcasts. Domestic Games revenue grew 17%, primarily driven by "Delta Force," "Valorant PC," "Valorant Mobile," and "Roco Kingdom: World." International Games revenue was down 1%, although up 4% in constant currency terms, as revenue growth from "Wuthering Waves" and "Valorant PC" was offset by revenue decreases from two Supercell games. For communications and social networks, Video Accounts total time spent grew over 20% in Q2, benefiting from enriched content supply, upgraded interactivity, and the introduction of a new multivariable content ranking system.

Speaker #1: We can enhance Tencent Music's resilience, deepen the content supply relationship between China literature and Shimalaya, and provide users with new content formats including audiobooks and podcasts.

Speaker #1: Domestic games revenue grew 17% primarily driven by Delta Force, Valorant PC, Valorant Mobile, and Roku Kingdom World. International games revenue was down 1%, although up 4% in constant currency terms as revenue growth from Wuthering Waves and Valorant PC was offset by revenue decreases from two Supercell games.

Speaker #1: For communications and social networks, video accounts' total time spent grew over 20% in the second quarter, benefiting from enriched content supply, upgraded interactivity, and the introduction of a new multivariable content ranking system.

Speaker #1: We've added content that appeals to younger users through IP partnerships with game studios, music labels, TV shows. And we've provided new revenue sharing opportunities for creators expanding the population of creators that generate direct revenue from within the video accounts.

James Mitchell: We've added content that appeals to younger users through IP partnerships with game studios, music labels, TV shows. We've provided new revenue-sharing opportunities for creators, expanding the population of creators that generate direct revenue from within the Video Accounts. Mini Shops GMV increased, within which GMV generated from Weixin centralized e-commerce gateway page grew significantly. For Mini Shops merchants, we introduced marketing tools such as lucky draws, helping them to enhance brand awareness and drive product discovery. For Mini Shops consumers, we enhanced rewards for repeat shoppers to increase customer life cycles, and thus customer lifetime value to merchants. On domestic games, "Delta Force" achieved lifetime high average DAU in Q2, driven by the Burst Fest campaign, the game's first professional esports final, and a global 20 versus 20 tournament.

Speaker #1: Mini shops GMV increased within which GMV generated from waging centralized e-commerce gateway page grew significantly. For mini shops merchants, we introduced marketing tools such as lucky draws, helping them to enhance brand awareness and drive product discovery.

Speaker #1: And for Mini Shops consumers, we enhanced rewards for repeat shoppers to increase customer life cycles and, thus, customer lifetime value to merchants. On domestic games, Delta Force achieved lifetime-high average DAU in the second quarter, driven by the Burst Fast campaign, the game's first professional esports final, and a global 20-versus-20 tournament.

Speaker #1: In terms of production, the Delta Force team have integrated AI across multiple workflows including using data agents for performance analysis and the Hunyuan 3D model for asset generation.

James Mitchell: In terms of production, the "Delta Force" team have integrated AI across multiple workflows, including using data agents for performance analysis and the Hunyuan 3D model for asset generation. "Valorant PC" also achieved lifetime high average DAU in Q2, benefiting from the Skirmish: Ascension mode with round progressive weapons and the Summit map with droppable walls. The game expanded its reach via influencer collaborations on the ground city events and promotions in over 10,000 Internet cafes. Among new games, "Roco Kingdom: World" ranked fifth by average DAU and eighth by gross receipts across all mobile games released industry-wide in Q2, making it the highest-ranked new title released year to date. The game has maintained a rapid content delivery cadence since launch, adding 100 creatures and expanding the map with seven new regions.

Speaker #1: Valorant PC also achieved lifetime high average DAU in the second quarter benefiting from the skirmish ascension mode with round progressive weapons and the summit map with droppable walls.

Speaker #1: The game expanded its reach through influencer collaborations, on-the-ground city events, and promotions in over 10,000 internet cafes. Among new games, Roku Kingdom World ranked fifth by average DAU and eighth by gross receipts across all mobile games released industry-wide in the second quarter.

Speaker #1: Making it the highest-ranked new title released year to date. The game has maintained a rapid content delivery cadence since launch, adding 100 creatures and expanding the map with seven new regions.

Speaker #1: On July the 9th, we released runaway evolution. This game is adapted from Rust, a survival open world crafting game on PC that has generally ranked among the top 20 games on Steam by concurrent users for the past eight years, thanks to its unique high risk, high reward gameplay in which players compete to outlast the other players on their server in one week competitive sprints.

James Mitchell: On 9 July, we released "Runaway Evolution." This game is adapted from "Rust," a survival open-world crafting game on PC that has generally ranked among the top 20 games on Steam by concurrent users for the past 8 years, thanks to its unique high-risk, high-reward gameplay in which players compete to outlast the other players on their server in one-week competitive sprints. "Runaway Evolution" seeks to tailor this gameplay for China market preferences by availability on mobile as well as PC devices, and via a sandbox safe zone for new players. Among our international games, "League of Legends" DAU increased year-on-year in Q2, primarily driven by the ARAM Mayhem mode. We launched "League Classic," a nostalgia mode that reengages longtime fans by recreating early-era gameplay with pre-reworked champions, classic runes, and the original Summoner's Rift map layout.

Speaker #1: Runaway evolution seeks to tailor this gameplay for China market preferences PC devices and via a sandbox safe zone for new players. Among our international games, League of Legends DAU increased year on year in the second quarter primarily driven by the ARAM Mayhem mode.

Speaker #1: We launched League Classic, a nostalgia mode that re-engages longtime fans by recreating early-era gameplay with pre-reworked champions, classic runes, and the original Summoner's Rift map layout.

Speaker #1: Warframe's DAU grew year on year, and gross receipts achieved a lifetime high in this quarter, benefiting from a new wolf-themed Prime Warframe and a new storyline, Jade Shadows Constellations.

James Mitchell: Warframe's" DAU grew year-on-year, and gross receipts achieved a lifetime high in this quarter, benefiting from a new wolf-themed Voruna Prime Warframe and a new storyline, Jade Shadows: Constellations. "Arrows: Puzzle Escape," a maze-clearing game developed by Miniclip subsidiary Lessmore, was the most downloaded mobile game globally in Q2. Arrows' success demonstrates that the innovative capabilities of Miniclip's family of studios, supported by Miniclip's publishing expertise, can together pioneer and break out leaders in new genres of casual games. Arrows monetizes via in-app advertising, so we report its revenue in our marketing services segment rather than the international games sub-segment. Adjusted to include Arrows and other in-app advertising game revenue in the prior year and current periods, our international games year-on-year revenue growth would have been 4 percentage points faster than the disclosed figures.

Speaker #1: Arrows Puzzle Escape Amaze Clearing Game, developed by Miniclip subsidiary Lesmor, was the most downloaded mobile game globally in the second quarter. Arrows' success demonstrates that the innovative capabilities of Miniclip's family of studios, supported by Miniclip's publishing expertise, can together pioneer and break out leaders in new genres of casual games.

Speaker #1: Arrows monetizes via in-app advertising, so we report its revenue in our marketing services segment rather than the international games subsegment. Adjusted to include Arrows and other in-app advertising game revenue in the prior year and current periods, our international games year-on-year revenue growth would have been four percentage points faster than the disclosed figures.

Speaker #1: For marketing services, revenue grew 22% year-on-year to RMB 44 billion, driven by higher eCPM and impressions. Most major categories increased their marketing spending with us, including e-commerce, internet services, and local services.

James Mitchell: For marketing services, revenue grew 22% year-on-year to RMB 44 billion, driven by higher eCPM and impressions. Most major categories increased their marketing spending with us, including e-commerce, internet services, and local services. We upgraded AI Marketing Plus end-to-end execution capabilities to better support closed-loop Mini Shop and Mini Drama advertisers. For example, AI Marketing Plus now enables Mini Shop owners to automatically select products for promotion, generate product-relevant ad creatives, and then run smart bidding to buy inventory for those creatives. We significantly scaled up the parameters of our advertising AI recommendation system to capture user interest with greater granularity and thus improve ad conversion rates. Video Accounts ad impressions grew rapidly year-on-year, driven by higher video views and ad load, although ad loads remain well below the short video industry average. Mini Programs attracted increasing marketing spend from Mini Drama and Mini Game Studios.

Speaker #1: We upgraded AI marketing plus end-to-end execution capabilities to better support closed loop mini shop and mini drama advertisers. For example, AI marketing plus now enables mini shop owners to automatically select products for promotion generate product relevant ad creatives and then run smart bidding to buy inventory for those creatives.

Speaker #1: We significantly see out of the parameters of our advertising AI recommendation system to capture user interest with greater granularity and thus improve ad conversion rates.

Speaker #1: Video accounts ad impressions grew rapidly year on year driven by higher video views and ad load although ad loads remained well below the short video industry average.

Speaker #1: Mini programs attracted increasing marketing spend from mini drama and mini game studios. For fintech and business services, segment revenue was RMB 60 billion, up 9%.

James Mitchell: For Fintech and Business Services, segment revenue was RMB 60 billion, up 9%. Fintech Services revenue grew year-on-year, driven by increases in commercial payment, wealth management, and consumer loan services. For commercial payment, the number of transactions grew year-on-year, while the decline in value per transaction narrowed. For wealth management, aggregated customer assets increased year-on-year, benefiting from the popularity of automated investment strategies and thematic index funds. Within Business Services, while we are still working through capacity constraints, our cloud revenue growth rate accelerated from high teens percentage year-on-year in Q1 to low 20s percentage in Q2, benefiting from AI-related demand, international expansion, and increased usage and pricing for general cloud services. AI-related demand translated into increased revenue across GPU rental, Model-as-a-Service, and WorkBuddy and CodeBuddy token usage. Our international cloud business expanded rapidly.

Speaker #1: Fintech services revenue grew year on year, driven by increases in commercial payment, wealth management, and consumer loan services. For commercial payment, the number of transactions grew year on year, while the decline in value per purchase action narrowed.

Speaker #1: For wealth management, aggregated customer assets increased year on year benefiting from the popularity of automated investment strategies and thematic index funds. Within business services, while we're still working through capacity constraints, our cloud revenue growth rate accelerated from high teams percentage year on year in the first quarter to low 20s percentage in the second quarter benefiting from AI related demand internationally expansion and increased usage and pricing for general cloud services.

Speaker #1: AI related demand translated into increased revenue across GPU rental, model as a service, and work by the encode by the token usage. Our international cloud business expanded rapidly using skills developed with code body is enabling us to conduct customer cloud migrations over to Tencent Cloud faster than we could in the past.

James Mitchell: Using skills developed with CodeBuddy is enabling us to conduct customer cloud migrations over to Tencent Cloud faster than we could in the past, for example, on behalf of a leading telecom company in Indonesia. I will pass to John.

Speaker #1: For example, on behalf of the leading telecom company in Indonesia. And now our pass to John.

Speaker #2: Thank you, James. For the second quarter of 2026, total revenue was RMB 204.8 billion, up 11% year on year. Gross profit was RMB 118.4 billion, up 13% year on year.

John Lo: Thank you, James. For Q2 2026, total revenue was RMB 204.8 billion, up 11% year-on-year. Gross profit was RMB 118.4 billion, up 13% year-on-year. Operating profit was RMB 67.3 billion, up 12% year-on-year. Interest income was RMB 4.2 billion, up 2% year-on-year. Finance costs were RMB 3 billion, compared with RMB 3.9 billion in the same period last year, reflecting favorable forex movements and lower interest expenses due to lower average interest rates. Our share of losses of associate and joint venture was RMB 10 billion for Q2 2026, primarily reflecting our share of the fair value adjustment recognized by an unlisted investee from the valuation of this issued convertible redeemable preferred shares arising from increased valuation of the investee, which was excluded from our non-IFRS profits.

Speaker #2: Operating profit was 67.3 billion renminbi, up 12% year on year. Interest income was 4.2 billion renminbi, up 2% year on year. Finance costs were 3 billion renminbi, compared with 3.9 billion renminbi in the same period last year, reflecting favorable forex movements and lower interest expenses due to lower average interest rates.

Speaker #2: Our share of losses of associate and joint venture was 10 billion renminbi for the second quarter of 2026. Primarily reflecting our share of the fair value adjustment recognized by an unlisted investee from revaluation of this issue convertible redeemable preferred shares arising from increased valuation of the investee.

Speaker #2: Which was excluded from our non-IFRS profits. On a non-IFRS basis, our share of profit of associates and joint venture for this quarter was 6.4 billion renminbi, compared with share of profits of 6.3 billion renminbi in the same period last year.

John Lo: On a non-IFRS basis, our share profit of associates and joint venture for this quarter was RMB 6.4 billion, compared with share profits of RMB 6.3 billion in the same period last year. Income tax expense increased by 3% year-on-year to RMB 11.7 billion. On non-IFRS financial figures, operating profit was RMB 75.6 billion, up 9% year-on-year. Operating profit excluding new AI products was RMB 86.1 billion, up 19% year-on-year. Net profit attributable to equity holders was RMB 68.4 billion, up 9% year-on-year. Diluted EPS was RMB 7.433, up 9% year-on-year. Moving on to gross margins for Q2. Overall gross margin was 58%, up 1 percentage point year-on-year. By segment, VAS gross margin increased by 4 percentage points year-on-year to 64%, driven by a favorable revenue mix shift towards high-margin internally developed gains.

Speaker #2: Income tax expense increased by 3% year-on-year to RMB 11.7 billion. On non-IFRS financial figures, operating profit was RMB 75.6 billion, up 9% year-on-year.

Speaker #2: Operating profit excluding new AI products was 86.1 billion renminbi, up 19% year on year. Net profit attributable to equity holders was 68.4 billion renminbi, up 9% year on year.

Speaker #2: Diluted BPS was 7.433 renminbi, up 9% year on year. Moving on to gross margins for Q2, overall gross margin was 58%, up 1 percentage point year on year.

Speaker #2: By segment, last gross margin increased by 4 percentage points year-on-year to 64%, driven by a favorable revenue mix shift towards high-margin, internally developed gains.

Speaker #2: Marketing services gross margin was 57%, down 0.3 percentage points year-on-year, as higher revenue supported by enhancements to our AI-driven marketing capabilities largely offset higher costs, including depreciation and operating costs associated with expanding our AI infrastructure.

John Lo: Marketing Services gross margin was 57%, down 0.3 percentage points year-on-year, as higher revenue supported by enhancement to our AI-driven marketing capabilities, partially offset higher costs, including depreciation and operating costs associated with expanding our AI infrastructure to improve ads and content recommendation. Fintech and Business Services gross margin was 52%, broadly stable year-on-year. On operating expenses, selling and marketing expenses were RMB 11.9 billion, up 26% year-on-year due to higher marketing spend to support our games business and to drive adoption of our AI-native products. R&D expenses rose by 35% year-on-year to RMB 27.2 billion, primarily reflecting higher R&D spend to support Hunyuan model enhancements, Huaxin AI initiatives, and development of AI capabilities across our products and services. G&A, excluding R&D expenses, decreased by 1% year-on-year to RMB 11.5 billion.

Speaker #2: The improved ads and content recommendation. Fintech and business services gross margin was 52%, broadly stable year on year. On operating expenses, selling marketing selling and marketing expenses were 11.9 billion renminbi, up 26% year on year due to higher marketing spend to support our games business and to drive adoption of our AI native products.

Speaker #2: R&D expenses rose by 35% year on year to RMB 27.2 billion, primarily reflecting higher R&D spend to support one-year model enhancements, creation AI initiatives, and development of AI capabilities across our products and services.

Speaker #2: GNA excluding R&D expenses decreased by 1% year on year to 11.5 billion renminbi. At quarter end, we had approximately 116,000 employees up 4% year on year and 1% quarter on quarter.

John Lo: At quarter end, we had approximately 116,000 employees, up 4% year-on-year and 1% quarter-on-quarter, mainly driven by headcount additions to our games and our technology platform, including AI-related headcount. Our Q2 non-IFRS operating margin was 36.9%, down 0.6 percentage point year-on-year. Non-IFRS operating margin excluding new AI products was 42%, up 2.8 percentage points year-on-year. To conclude, I will highlight some key cash flow and balance sheet metrics. Operating CapEx was RMB51.8 billion, up 190% year-on-year or 66% quarter-over-quarter as we accelerated investments in AI infrastructure to support Hunyuan model enhancements with PAPI and coPAPI inference needs, Huaxin AI initiatives, and development of AI capabilities across our products and services, as well as to meet growing external demand for our cloud services. Non-operating CapEx was RMB1 billion.

Speaker #2: Mainly driven by headcount additions to our games and our technology platform, including AI related headcount. Our second quarter non-IFRS operating margin was 36.9%, down 0.6 percentage point year on year.

Speaker #2: Non-IFRS operating margin, excluding new AI products, was 42%, up 2.8 percentage points year on year. To conclude, I will highlight some key cash flow and balance sheet metrics.

Speaker #2: Operating capex was 51.8 billion renminbi, up 190% year on year or 66% quarter over quarter. As we accelerated investments in AI infrastructure to support.

Speaker #2: One year model enhancement with body and code body inference needs. Weight sheen AI initiatives and development of AI capabilities across our products and services.

Speaker #2: As well as to meet growing external demand for our cloud services. Non-operating capex was 1 billion renminbi. Free cash flow was negative 13.8 billion renminbi, reflecting large AI infrastructure capex and AI related prepayments.

John Lo: Free cash flow was negative RMB13.8 billion, reflecting large AI infrastructure CapEx and AI-related prepayments, as well as seasonally lower games gross receipts. Excluding the prepayments for compute procurement, free cash flow would have been RMB37.6 billion. Net cash position was RMB58.2 billion, compared with RMB146.9 billion as at 31 March 2026, reflecting capital expenditure payments of RMB59.3 billion and 2025 dividend payments of RMB41.6 billion made during the quarter. Thank you.

Speaker #2: As well as seasonally lower games gross receipts. Excluding the prepayments for compute procurement, free cash flow would have been 37.6 billion renminbi. Net cash position was 58.2 billion renminbi compared with 146.9 billion renminbi as at 31st of March 2026, reflecting capital expenditure payments of 59.3 billion renminbi and 2025 dividend payments of 41.6 billion renminbi made during the quarter.

Speaker #2: Thank you.

Speaker #1: Thank you, John. We shall now open the floor for questions. If you are dialing in by phone, please press five to raise a question and then press six to unmute yourself.

Wendy Huang: Thank you, John. We shall now open the floor for questions. If you are dialing in by phone, please press 5 to raise a question and then press 6 to unmute yourself. If you are accessing from the Tencent Meeting or VooV Meeting application, please click the raise hand button at the bottom. We will take one main question, up to one follow-up question each time. The first question comes from Robin Zhu from Bernstein. Robin, your line is open.

Speaker #1: If you are accessing from the Tencent meeting or group meeting application, please click the raise hand button at the bottom. We will take one man question up to one follow-up question each time.

Speaker #1: The first question comes from the Robin Joo from Bernstein. Robin, your line is open.

Speaker #3: Thanks, Wendy. Thanks, management. For the opportunity to ask a question. I guess if we look at your latest quarter's capex, 53 billion it's a step up from the previous quarter annualizes over 200 billion if we just multiply by four.

Robin Zhu: Thanks, Wendy. Thanks, management, for the opportunity to ask a question. I guess if we look at your latest quarter's CapEx, 53 billion. It is a step up from the previous quarter, annualizes over 200 billion if we just multiply by 4. How should we think about the D&A costs, the results from this? To what extent do you think this will be paid off from incremental revenues that comes as a result of your investments in AI? Or is this essentially eating into earnings into the next few quarters? I would love to hear your thoughts on the time lags involved when it comes to the payback cycle, especially if we include some of the R&D costs incurred as well. Thank you.

Speaker #3: How should we think about the DNA costs that results from this? To what extent do you think this will be paid off from incremental revenues that come as a result of your investments in AI?

Speaker #3: Or is this essentially eating into earnings over the next few quarters? I’d love to hear your thoughts on the timelines involved, especially regarding the payback cycle if we include some of the R&D costs incurred as well.

Speaker #3: Thank you.

Speaker #2: Thank you for the question, Robin. So I mean, given the surge in demands and therefore rental pricing for compute, we could recover the depreciation almost immediately by renting the compute out to third parties as many neo cloud businesses are doing.

John Lo: Thank you for the question, Robin. Given the surge in demand and therefore rental pricing for compute, we could recover the depreciation almost immediately by renting the compute out to third parties, as many neo cloud businesses are doing. We would then achieve a decent return in an immediate timeframe. However, in reality, we are playing a different game or executing a larger strategy in that we are allocating a very substantial proportion of the new compute to building our own models to state-of-the-art status, and also to deploying, popularizing, and bringing our own AI applications to market leadership in China. Our belief is that by providing the superior intelligence that we can achieve through state-of-the-art models, through market-leading AI applications, that superior intelligence we can then convert into superior economic returns over the longer term. For example, by selling tokens through the work by the application.

Speaker #2: And we would then achieve a decent return in an immediate timeframe. However, in reality, we're playing a different game or executing a larger strategy in that we're allocating a very substantial proportion of the new compute to building our own models to state-of-the-art status.

Speaker #2: And also to deploying popularizing and bringing our own AI applications to market leadership in China. And our belief is that by providing the superior can achieve through state of the art models, through market leading AI applications, that superior intelligence, we can then convert into superior economic returns over the longer term.

Speaker #2: For example, by selling tokens through the Work Body application. So that's the path we've chosen.

John Lo: That is the path we have chosen.

Speaker #4: So just to elaborate a little bit more on this, right? So I think at the time being, you can actually sort of look at the Tencent businesses and break it into two businesses.

Martin Lau: Just to elaborate a little bit more on that. I think at the time being, you can actually look at the Tencent businesses and break it into two business. One is actually the existing core franchises, which actually generates solid growth and also with quite a bit of operating leverage. That is the high-quality growth track that we have been building, and we continue with that. Then there is another new AI native business that we are actually building. The new AI native business would involve, as James said, our own model as well as new applications that we are building and also a corresponding compute infrastructure. The financials that you should look at there is the revenue and profit in relation to our core existing business. We do separately disclose the investments in our AI native business, as an operating line.

Speaker #4: One is actually the existing, or franchises, which actually generate solid growth and also with quite a bit of operating leverage. That's the high-quality growth threat that we have been building, and we continue with that.

Speaker #4: And then there's another new AI native business that we're actually building. And the new AI native business would involve as James said, our own model as well as new applications that we're building and also sort of a corresponding compute infrastructure.

Speaker #4: And the financials that you should look at, there is the revenue and profit in relation to our existing business and we do separate disclose the investments in our AI native business as an operating line.

Speaker #4: And then, when you look at the CapEx, I would say the CapEx would be divided into two parts too, right? One part is really in relation to our existing business, which you can just like in the past, right?

Martin Lau: Then when you look at the CapEx, I would say the CapEx will be divided into two parts too, right? One part is really in relation to our existing business. Just like in the past, right? We can just say, oh, this is the free cash flow in which we generate operating cash flow, and there is a CapEx in relation to that. That part of the business still very cash flow generative. Then there is another set of CapEx which is related to the new AI native business, which is essentially a lump sum that we need to invest in order to get our compute for model training, in order to prepare for inference needs, and in order to also order some more for building our AI compute and AI cloud business. That is essentially what it is.

Speaker #4: You can just say, oh, this is the free cash flow in which we generate operating cash flow and there is a capex in relation to that.

Speaker #4: And that part of the business still very, very cash flow generative. And then there is another set of capex, which is related to the new AI native business, which is essentially a lump sum that we need to invest in order to get our compute for model training in order to prepare for inference needs.

Speaker #4: And in order to also order some more for building our AI compute and AI cloud business. So that's essentially what it is. And the reason we're actually investing in all these compute is that we need that in order to essentially get the business kick started.

Martin Lau: The reason we are actually investing in all these compute is that we need that in order to essentially get the business kick-started. At the same time, when we make the investment, there is clear upside that we are seeing because our model is doing well, our new applications is doing well, and we also have a lot of demand for compute. Today, if we can actually allocate the compute toward leasing on the Tencent Cloud, we would actually generate a lot more revenue and would generate significant return from the CapEx. As a matter of fact, for the prepayment and for some of the compute orders that we have made just a couple of months ago, today, we can actually sell it at more than 30% profit compared to the price that we paid just a few months ago.

Speaker #4: And at the same time, when we make the investment, there's clear upside that we're seeing because our model is doing well. Our new applications are doing well.

Speaker #4: And we also have a lot of demand for compute. Today, if we can actually allocate the compute toward leasing on the Tencent cloud, we've actually sort of generate a lot more revenue.

Speaker #4: And we'll generate significant return from the capex. As a matter of fact, for the prepayment and for some of the compute orders that we have made just a couple of months ago, today we can actually sell it at more than 30% profit.

Speaker #4: Compared to the price that we paid just a few months ago. But we believe if we use this compute for building our own model and building our application and then allocating the compute for rental in that order, over time we'll build a very significant AI native business and that will be hugely profitable as well as highly cashed as well as return generative for Tencent.

Martin Lau: But we believe if we use this compute for building our own model and building our application and then allocating the compute for rental in that order, over time, we will build a very significant AI native business and that will be hugely profitable, as well as highly cashed, as well as return generated for Tencent. That is the way we think about the business right now.

Speaker #4: So that's the way we think about the business right now.

Speaker #3: Got it. Thank you. And if I may have a follow up just on work body I'd love to hear your thoughts on every AI lab is essentially incentivized to develop their own harness app of some kind.

Robin Zhu: Got it. Thank you. If I may have a follow-up just on WorkBuddy. Love to hear your thoughts on, every AI lab is essentially incentivized to develop their own harness app of some kind, and your thoughts on how the market breaks down between first and third-party harness apps, how you would like to set up WorkBuddy to compete against these first-party harnesses and whether WorkBuddy, in your mind, is a piece of enterprise software that sits next to Tencent Meeting, Docs, or is this a new platform play that essentially becomes a marketplace for AI in the future? Thanks.

Speaker #3: And your thoughts on how the market breaks down between first- and third-party harness apps, how you would like to set up Work Body to compete against these first-party harnesses, and whether Work Body, in your minds, is a piece of enterprise software that sits next to Tencent Meeting Docs, or is this a new platform play that essentially becomes a marketplace? Thanks.

Speaker #4: Yeah. Well, I think it is indeed a new platform. It's a very flexible workspace for AGN to AI. The core purpose is actually that it will solve all the productivity needs of office workers and of all kinds of people who engage in their own businesses, right?

Martin Lau: Well, I think it is indeed a new platform that. It is a very flexible workspace for agentic AI. The core purpose is actually it will solve all the productivity needs of office workers and of all kinds of people who engage in their own businesses, right? One-person companies and the like. Below that, there will be a harness which actually helps the users to make use of the capability of different models to solve the agentic problems of the users. Over time, there will be many models serving the users through WorkBuddy. There will be many skills developed over time by all kinds of different developers. The purpose is actually solving productivity problems, and then the platform itself would make use of all kinds of different tools and models available to do that. Then, of course, we are the orchestrator, right?

Speaker #4: One person companies and the like. And below actually sort of helps to helps the users to make use of the capability of different models to solve the AGN problems of the users.

Speaker #4: And over time, there will be many models serving the users through work body that will be many skills developed over time by all kinds of different developers.

Speaker #4: And the purpose is actually solving productivity problems and then the platform itself would make use of all kinds of different tools and models available to do that.

Speaker #4: And of course, we are the orchestrator, right? So we can actually choose the right model and choose the right skills to help users solve the problems.

Martin Lau: We can actually choose the right model, choose the right skills to help users solve the problems. We choose that to make sure that the work is done perfectly, but at the same time, it will be done also very economically, right? Hunyuan will be one of the models that will be provided by WorkBuddy. At the same time, if you can actually solve a lot of the user problems, right, it is quite effective, then Hunyuan would actually be one of the main models within WorkBuddy. It would not be the only model. Thank you, Raj.

Speaker #4: And we choose that to make sure that the work is done perfectly, but at the same time, it will also be done very economically, right?

Speaker #4: And when you will be one of the models that will be provided by Workbody, but at the same time, if you can actually solve a lot of the user problems, right?

Speaker #4: And it's quite effective, then, when you would actually be one of the main models within the work body. But it will not be the only model.

Speaker #3: Thank you very much.

Robin Zhu: Thank you, Raj.

Speaker #5: Thanks, Robin. We will take the next question from Kenneth Phone from UBS.

Wendy Huang: Thanks, Robin. We will take the next question from Kenneth Fong from UBS.

Speaker #6: Hi. Good evening, management and thanks for taking my question. I have a question regarding the Xiaowei development. So could management share any preliminary feedback or challenges from the testing phase of Xiaowei?

Kenneth Fong: Hi. Good evening, management, and thanks for taking my question. I have a question regarding the Xiaowei development. Could management share any preliminary feedback or challenges from the testing phase of Xiaowei? From a commercial standpoint, how should we evaluate the net monetization potential, specifically as agents simplify the transaction path, we worry that it may just be shifting the existing volume away from traditional user self-performed transaction in Mini Program over to the agents, which carry a higher computing cost without a meaningfully higher net new GTV. Furthermore, would the shortened user transaction journey in Xiaowei risk also lowering the high margin ad impression inventory as well? Thank you very much.

Speaker #6: And from a commercial standpoint, how should we evaluate the net monetization potential? Specifically, as agents simplify the transaction path, we worry that it may just be shifting the existing volume away from traditional user self-performed transaction in many programs.

Speaker #6: Over to the agents, which carry a higher computing cost. Without a meaningfully higher net new GTV. And furthermore, would the short-term user transaction journey in Xiaowei risk also lowering the high margin ad impression inventory as well?

Speaker #6: Thank you very much.

Speaker #4: Well, I think all the risks that you said would not be relevant. Because we believe when AI enable the ways in ecosystem to be more intelligent and it can actually sort of help users to execute transactions, explore content, and manage their daily life with a lot of AI, right?

Martin Lau: Well, I think all the risks that you said were not irrelevant. We believe when AI enable the Weixin ecosystem to be more intelligent and it can actually sort of help users to execute transactions, explore content, manage their daily life with a lot of AI, right? Then the Weixin ecosystem, which is already very rich and powerful, will become even more useful to the users, right? If you imagine, the time when QQ was a communication and social tool in the PC stage, then when we get into the mobile age, Weixin appears and Weixin essentially sort of the ecosystem magnified QQ's value by more than 10x, right? Because it is enabled in the mobile age and it becomes mobile first.

Speaker #4: Then the AI, the ways in the ecosystem, which is already very rich and powerful, will become even more useful to users, right? So if you imagine the time when QQ was a communication and social tool in the PC stage, and then when we got into the mobile age, then Weixin appears—and Weixin essentially, sort of, the ecosystem magnified QQ's value by more than 10x, right?

Speaker #4: Because it's enabled in the AI in the mobile age and it becomes mobile first. So when we look at AI, we believe there's another huge opportunity for the ways in ecosystem to be first enabled by AI and over time it will be AI first application and ecosystem.

Martin Lau: So when we look at AI, we believe there is another huge opportunity for the Weixin ecosystem to be first enabled by AI, and over time, it will be AI first application and ecosystem. When that happens, users would have a lot of great experiences. Right now, you actually sort of have to type and you have to sort of navigate through clicks. In the future, if you just tell Xiaowei one instruction, then Xiaowei can go off and help you execute a transaction and execute your instruction, and that would be an incredible experience for the users. It would also be an incredible empowerment for the entire ecosystem.

Speaker #4: And when that happens, users would have a lot of great experiences like right now you actually sort of have to type and you have to sort of navigate through clicks in the future if you just tell Xiaowei, one instruction and then Xiaowei can go off and help you execute a transaction.

Speaker #4: And execute your instruction, and that will be an incredible experience for the users. It would also be an incredible empowerment for the entire ecosystem.

Speaker #4: So we believe if we can deliver that experience, if we can control the cost of that delivery and if you look at the design of WLM is actually for privacy, for cost efficiency, and for making sure that it can execute within the ways in environment all the needs of the users, right?

Martin Lau: We believe if we can deliver that experience, if we can control the cost of that delivery, and if you look at the design of WeLM, it is actually for privacy, for cost efficiency, and for making sure that it can execute within the Weixin environment all the needs of the users, right? If we can do that, then we can really empower Weixin for the AI age under controllable cost. When that happens, Weixin's ecosystem would expand, and that would translate into a lot of value just based on the current monetization mechanisms within Weixin. I think that is the future that we are seeing. With the launch of the prototype, we grow more and more confident about that will be happening.

Speaker #4: And if we can do that, then we can really empower Waisin for the AI age under controllable cost. And when that happens, the Waisin ecosystem would expand, and that would translate into a lot of value just based on the current monetization mechanisms within Waisin.

Speaker #4: And I think that's the future that we're seeing. And with the launch of the prototype, we grow more and more confident about that will be happening.

Speaker #6: Thank you, Martin. Another follow-up question on the AI cloud with domestic API token prices favoring rapid amortizations. So and also China cloud market remains structurally price sensitive.

Kenneth Fong: Thank you, Martin. I have a follow-up question on the AI cloud. With domestic API token prices face very rapid commoditizations. China cloud market remains structurally price sensitive. How do we think about the margin profile of Tencent AI Cloud currently compared to, say, PaaS and SaaS offering? As this gradually scale up as AI adoption scale, how should we also think about the margin progression going forward? Thank you.

Speaker #6: So how do we think about the margin profile of Tencent AI cloud currently compared to say us and past offering? And as this gradually scale up as AI abduction scale, so how should we also think about the margin progression going forward?

Speaker #6: Thank you.

Speaker #2: Well, it is true that domestic token price is low, but the domestic token manufacturing costs are also extremely low. And I think much lower than widely perceived or externally estimated.

James Mitchell: Well, it is true that domestic token prices are low, but the domestic token manufacturing costs are also extremely low, I think much lower than widely perceived or externally estimated. The token business, it can be positive gross margin at these low token prices because the cost is low. If you look at the gross margin for the paying users of WorkBuddy, or you look at the gross margin for our models of service, then the gross margins today are already comparable to the gross margins for Tencent Cloud overall. Of course, WorkBuddy in aggregate has a lower gross margin because there is a proportion of free users whom we are subsidizing to drive market share and market growth. But on the paying users, we are generating a pretty good gross margin right now. On your broader concern, it is true also that the China cloud market is price competitive.

Speaker #2: So the token business, it can be positive gross margin at these low token prices because the cost is low. And if you look at the gross margin for the paying users of work body, or you look at the gross margin for our models of service, then the gross margins today are already comparable to the gross margins for Tencent cloud overall.

Speaker #2: Of course, work body in aggregate has a lower gross margin because there's a proportion of free users whom we're subsidizing to drive market share and market growth.

Speaker #2: But on the paying users, we're generating a pretty good gross margin right now. And on your broader concern, it is true also that the China cloud market is price competitive.

Speaker #2: But that environment has changed a great deal in the last several months, as input costs—particularly for memory—have gone up. So we've been increasing the prices we charge to our customers.

James Mitchell: But that environment has changed a great deal in the last several months as the input costs, particularly for memory, have gone up. So we have been increasing the prices we charge to our customers. We increased prices across the board in May for Tencent Cloud. Beyond those headline price increases, we have also been more substantially reducing discounts. So the overall pricing environment in cloud in China is not as difficult as it has been in the past. Thank you.

Speaker #2: We are increased prices across the board in May for Tencent cloud. And beyond those headline price increases, we've also been more substantially reducing discounts.

Speaker #2: So the overall pricing environment in cloud in China is not as difficult as it's been in the past. Thank you.

Speaker #1: Thank you, Kenny. We will take the next question from Ronald Kwon from Goldman Sachs.

Wendy Huang: Thank you, Kenneth. We will take the next question from Ronald Keung from Goldman Sachs.

Speaker #7: Thanks, Boni, Martin, James, John, and Wendy. So two questions. I think first on the Huan Yuan model. Just want to hear after the progress of Huan Yuan 3, which is on cost efficiency, I would say in very good in agents.

Ronald Keung: Thanks. Pony, Martin, James, John, and Wendy. So two questions. First on the Hunyuan model. Just want to hear after the progress of Hunyuan 3, which is on cost efficiency, I would say, and very good in agents. Where will Hunyuan 4 differentiate itself as we look into a, let us say, 3 trillion parameter size class is looking more crowded in the next few months. So which category are we looking or which segment or differentiation are we thinking for Hunyuan 4? A second question is on the CapEx and the focus on our AI initiatives. Looking at some of the US peers where there has been shift in strategy on the hyperscaler business.

Speaker #7: Then where will Huan Yuan 4 differentiate itself as we look into a, let's say, 3 trillion perimeter size class is looking more crowded in the next few months?

Speaker #7: So which category are we looking or which segment or differentiation are we thinking for Huan Yuan 4? And then a second question is on the CapEx and the focus on our AI initiatives.

Speaker #7: But looking at some of the US peers where there has been a shift in strategy on the hyperscaler business, I just want to hear what stage or timeline we think we may focus more on cloud as a potential high-ROI business that is worth prioritizing more CapEx on.

Ronald Keung: Just want to hear what stage or timeline that we think we may focus more on cloud as a potential high ROI business that is worth prioritizing more CapEx on, and some similarities and differences that we see for Tencent Cloud versus how US peers have shifted their focus more from applications to cloud for some of our peers. So two questions. Thank you.

Speaker #7: And some similarities and differences that we see for Tencent cloud versus what how US peers have shifted that focus more from applications to cloud for some of our peers to questions.

Speaker #7: Thank you.

Speaker #4: If you look at the Huan Yuan 3, right, Huan Yuan 3 is a very small model, even in today's terms, but it's actually sort of very widely used right now.

Martin Lau: If you look at the Hunyuan 3, Hunyuan 3 is a very small model, even in today's terms, but it is actually very widely used. I think there are a number of characteristics of Hunyuan 3, which is it actually has the capability of matching or beating much larger models. That is one. Two is, it is actually focused on use cases rather than just benchmark beating. As a result, in real life, it has become much more useful than a lot of models of the same size or even bigger size. We believe that is a principle that we will be applying to Hunyuan 4 as well. When Hunyuan 4 comes out, it will be a bigger model, and it will be able to beat models of bigger size. It would also be extremely useful and more useful than Hunyuan 3.

Speaker #4: So I think there are a number of characteristics of Huan Yuan 3, which is it actually has the capability of beating the matching or beating much larger models.

Speaker #4: That's one. And two is it's actually focused on use cases rather than just benchmark beating. And as a result, in real life, it has become much more useful than a lot of models of the same size or even bigger size.

Speaker #4: We believe that's a principle that we will be applying to Huan Yuan 4 as well. So when Huan Yuan 4 comes out, it will be a bigger model.

Speaker #4: And it will be able to beat models of bigger size. And it will also be extremely useful and more useful than Huan Yuan 3.

Speaker #4: And we believe that would actually take us into the next stage of being able to provide much better intelligence to a lot of our users.

Martin Lau: We believe that would actually take us into the next stage of being able to provide much better intelligence to a lot of our users. Bear in mind that we also have products which are co-designing with a model. When Hunyuan 4 comes around, the products that would be using Hunyuan 4 would actually become even more powerful and even more useful than what they are today. That would actually provide a very significant lift for the products that it is powering. I think that is sort of the path, and Hunyuan 4 is only another stop. Then we will be upgrading to Hunyuan 5. As we continue to progress, we will be approaching SOTA, and at some point in time, we will definitely be able to reach SOTA.

Speaker #4: And bear in mind that we also have products which are co-designing with the models. So when Huan Yuan 4 comes around, the products that would be using Huan Yuan 4 would actually become even more powerful.

Speaker #4: And even more useful than what they are today. And that would actually provide a very significant lift for the products that it's powering. So I think that's sort of the path.

Speaker #4: And Huan Yuan 4 is only sort of another stop, right? And that will be sort of upgrading to Huan Yuan 5. So as we continue to progress, we will be approaching Sota and at some point in time, we'll definitely sort of be able to reach Sota.

Speaker #4: And once we are there, we would also have a lot of models of different sizes that will be able to solve different kinds of user problems at the different level of model and cost efficiency.

Martin Lau: Once we are there, we would also have a lot of models of different sizes that will be able to solve different kinds of user problems at the different level of model and cost efficiency. At the same time, we have multiple models that can be used for co-design with our different products, and that would help us to make the products feature-rich and help to make the products powerful as well as the speed of execution will be fast. I think that is what we envision Hunyuan 4 and then subsequently Hunyuan 5 to be like.

Speaker #4: And at the same time, we have multiple models that can be used for co-design with different products. And that would help us to make the products feature-rich and help to make the models and the products powerful, as well as improve the speed of execution.

Speaker #4: So I think that's what we envision Huan Yuan 4 and then subsequently Huan Yuan 5 to be like.

Speaker #2: And in terms of your second question about allocating CapEx between different use cases, including Tencent Cloud, the immediate primary use case for the CapEx is for training bigger and better Huan Yuan models in the coming months, as Martin discussed.

James Mitchell: In terms of your second question about allocating CapEx between different use cases, including Tencent Cloud. The immediate primary use case for the CapEx is for training bigger and better Hunyuan models in the coming months, as Martin discussed. An important secondary use case is providing inference for the use of Hunyuan models as well as DeepSeek and other models behind WorkBuddy. The intention of that WorkBuddy initiative is primarily to drive adoption of what we think is a strategically important application with critical feedback loops back to our model and our broader ecosystem. It also has the happy effect of generating revenue upfront. From an accounting perspective, the majority of the WorkBuddy spending by users is on subscriptions.

Speaker #2: But an important secondary use case is providing inference for the use of Huan Yuan models as well as deep seek and other models behind work body.

Speaker #2: And so the intention of that work body initiative is primarily to drive adoption of what we think is a strategically important application with critical feedback loops.

Speaker #2: Back to our model and our broader ecosystem. But it also has the happy effect of generating revenue up front. Now, from an accounting perspective, the majority of the work body spending by users is on subscriptions.

Speaker #2: And so similar to games and some of our other businesses, there's a lengthy time lag between the cash receipts coming to us from the users and those cash receipts translating into reported revenue.

James Mitchell: And so, similar to games and some of our other businesses, there is a lengthy time lag between the cash receipts coming to us from the users and those cash receipts translating into reported revenue. But we are seeing a substantial ramp in the cash receipts today, and that will translate into reported revenue growth for Tencent Cloud as we move through the year. Then toward the end of the year and into next year, we will also have sufficient GPU ASIC capacity to step up in terms of Tencent Cloud renting out bare metal GPU or providing Model-as-a-Service. But within those opportunities, renting out GPU Model-as-a-Service and then token production for WorkBuddy, we think that it is token production for WorkBuddy that carries the most enduring economic value to us, and that is why we are prioritizing it today. Thank you.

Speaker #2: But we are seeing a substantial ramp in the cash receipts today, and that will translate into reported revenue growth for Tencent Cloud as we move through the year.

Speaker #2: And then toward the end of the year and into next year, we'll also have sufficient GPU ASIC capacity to step up in terms of Tencent Cloud renting out bare metal GPU or providing models as a service.

Speaker #2: But within those opportunities, renting out GPU models as a service and then token production for work body we think that it is token production for work body that carries the most enduring economic value to us.

Speaker #2: And that's why we're prioritizing it today. Thank you.

Speaker #1: Thanks, Martin and James.

Ronald Keung: Thanks, Martin and James.

Speaker #5: Thank you. We will take the next question from Alicia Yap from Citigroup.

Wendy Huang: Thank you. We will take the next question from Alicia Yap from Citigroup.

Alicia Yap: Hi, good evening management. Thanks for taking my questions. Congrats on the solid results. First question is on the Xiaowei. Could management elaborate on your comment on the agent-to-agent transaction loop? Will this concept lead to the long-term visions for a fully autonomous agents ecosystem within the Weixin? Then management also highlight that you will explore the on-device inference for Xiaowei. So what are the challenges and the benefits of this approach? Then is this on-device inference approach another reasons why the proprietary WeLM models is more suitable, empowering the Xiaowei rather than the external model? Then a quick follow-up is on your marketing service revenue. So this quarter, the growth rate accelerated to 22%. Should we expect this ongoing upgrade of the attack and also this automated campaign to further support this growth momentum?

Speaker #6: Hi, good evening, management. Thanks for taking my questions, and congratulations on the solid results. My first question is on Xiaowei. Could management elaborate on your comment regarding the agent-to-agent transaction loop?

Speaker #6: Will this concept lead to the long-term visions for a fully autonomous agent's ecosystem within the Wexing? And then management also highlight that you will explore the on-device inference for Xiaowei?

Speaker #6: So what are the challenges and the benefits of this approach? And then is this on-device inference approach another reasons why the proprietary VLM models is more suitable in powering the Xiaowei rather than the external model?

Speaker #6: And then a quick follow-up is on your marketing services revenue. So this quarter, the growth rate accelerated to 22%. Should we expect this ongoing upgrade of the EdTech and also this automated campaign to further support this growth momentum?

Speaker #6: So any further future benefit that you would anticipate from the deeper integrations into your Huan Yuan 3 model? Thank you.

Alicia Yap: So any further future benefit that you would anticipate from the deeper integrations into your Hunyuan 3 model? Thank you.

Speaker #4: Yeah. So on the agent-to-agent transaction loop, I think we are envisioning a future in which a lot of users would be executing their instructions and over time transactions via Xiaowei and via agents, right?

Martin Lau: Yeah. On the agent-to-agent transaction loop, I think we are envisioning a future in which a lot of users would be executing their instructions and over time transactions, via Xiaowei and via agents, right? In the past, if you think about the Weixin ecosystem, users interacting with content, interacting with Mini Programs themselves. In the future, if they can actually send a complex instruction to an agent, then an agent can actually start helping the user to execute transactions. A lot of the Mini Programs, a lot of the merchants would actually also have agents, which over time can interact with the agent of the users. Longer term, there will be even user, each user has got an agent and they can actually interact with each other to execute transactions. So I think that essentially is what is possible for the future.

Speaker #4: And in the past, if you think about the Wexing ecosystem is users, interacting with content, interacting with mini programs themselves. And in the future, if they can actually send a complex instruction to an agent, then the agent can actually sort of start helping the user to execute transactions.

Speaker #4: And a lot of the mini programs a lot of the merchants would actually sort of also have agents which over time can interact with the agent of the users.

Speaker #4: And longer term, there will be even user each user has got an agent and they can actually interact with each other to execute transactions.

Speaker #4: So I think that essentially is what's possible for the future. And we're building the architecture for making that possible on a step-by-step basis. In terms of on-device inference, I think, number one, it will be happening maybe step-by-step, and it will be only over the long run that most of the inference will be happening on-device, right?

Martin Lau: We are building the architecture for making that possible on a step-by-step basis. In terms of on-device inference, I think it would, number one, be happening maybe step by step and it will be only over the long run that most of the inference will be happening on device, right? I think at some point in time, it is not hard to imagine some kind of inference will be actually happening on device, and some inference will be happening in the cloud. Over time, as the on-device compute becomes more and more powerful and as the model becomes more and more efficient, then you will have more inference happening on people's devices. I think that would be going back to the normal state of the computer industry.

Speaker #4: But I think, at some point in time, it's not hard to imagine some kind of inference will actually be happening on-device, and some inference will be happening in the cloud. Over time, as on-device compute becomes more and more powerful and as the model becomes more and more efficient, you'll have more inference happening on people's devices.

Speaker #4: And I think that would be going back to the normal state of the computer industry, right? If you think about the computer industry as well as the smartphone industry, right, most of the compute, right, which is CPU actually happens on-device.

Martin Lau: If you think about the computer industry as well as the smartphone industry, most of the compute, which is CPU, actually happens on device. The cloud actually is only responsible for a small part of the compute. In this initial phase of AI infrastructure, most of the compute, because it has to be very powerful, right? The problem of getting enough compute on device, getting it cheap enough, and also getting it power efficient enough has not happened yet. So that is why everything happens on the cloud. There will be a time in which more and more GPU capability will be put into everybody's phone and computer.

Speaker #4: And the cloud actually is only responsible for a small part of the compute. But in this initial phase of AI infrastructure, most of the compute—because it has to be very powerful, right?—is elsewhere.

Speaker #4: And the problem of getting enough compute on-device, getting it cheap enough, and also getting it power efficient enough has not happened yet. So that's why everything happens on the cloud.

Speaker #4: But there will be a time in which more and more GPU capability will be put into everybody's phone and computer and when that happens, then more and more inference will be happening on the device and there will be sort of going back to the time when it's actually the software, it's actually the model that becomes much more important and the return for running models and the return on running applications will be higher because the compute capex will be not just borne by the model company, but it will be borne across the ecosystem.

Martin Lau: When that happens, then more and more inference will be happening on the device and it will be sort of going back to the time when it is actually the software, it is actually the model that becomes much more important. The return for running models and the return on running applications will be higher because the compute CapEx will be not just borne by the model company, but it will be borne across the ecosystem. I think that would definitely happen at some point in time and we are building and preparing for that.

Speaker #4: And I think that will definitely happen at some point in time, and we're building and preparing for that.

Speaker #2: And on your marketing services question, our advertising revenue growth has ticked up and ticked down in the past, and it will continue to tick up and tick down in the future.

James Mitchell: On your marketing services question, our advertising revenue growth has ticked up and ticked down in the past, and it will continue to tick up and tick down in the future. I would not sort of straight line extrapolate anything. There is a number of reasons. One is that the sort of obverse of the comments I made about in-app advertising games being a drag on the international game segment revenue growth versus where it would otherwise have been, is that they did contribute about 2 percentage points to the advertising segment revenue growth this quarter. These in-app advertising games are a sort of new product for Tencent, to some extent, a new product for the world. So we do not have the same degree of clarity on what the growth trajectory will be for the in-app advertising game contribution as for our sort of conventional marketing services revenue.

Speaker #2: I wouldn't sort of straight-line extrapolate anything. And there are a number of reasons. One is that the sort of obverse of the comments I made about in-app advertising in games being a drag on the international games segment revenue growth versus where it would otherwise have been, is that they did contribute about 2 percentage points to the advertising segment revenue growth this quarter.

Speaker #2: And these in-app advertising games are sort of new product for Tencent to some extent a new product for the world. And so we don't have the same degree of clarity on what the growth trajectory will be for the in-app advertising game contribution as for our sort of conventional marketing services revenue.

Speaker #2: In addition, the China consumer and therefore advertising market remains choppy, and there are some economic or consumption headwinds. That may have an impact on advertising trends.

James Mitchell: In addition, the China consumer and therefore advertising market remains choppy and there are some economic or consumption headwinds that may have an impact on advertising trends. That said, we have been outperforming the overall China advertising market, and we are confident we will continue to do so by a substantial margin given the upside to us from deploying AI ad targeting, given the fact that engagements, especially for our key Video Accounts inventory, is increasing at a good rate, and given we are early in the evolution toward more closed-loop advertising that drives much higher ad pricing. Thank you.

Speaker #2: That said, we have been outperforming the overall China advertising market. And we're confident we'll continue to do so by a substantial margin. Given the upside to us from deploying AI ad targeting, given the fact that engagements especially for our key video accounts inventory is increasing at a good rate, and given we're early in the evolution toward more closed loop advertising that drives much higher ad pricing.

Speaker #2: Thank you.

Speaker #6: Thank you. We will take the next question from Alex Liu at Bank of America.

Wendy Huang: Thank you. We will take the next question from Alex Liu from Bank of America.

Speaker #1: Thank you for taking my questions. I have only one question. So we noted that Tencent has recently increased the buyback. The buyback activity started from May.

Alex Liu: Thank you for taking my questions. I have only one question. We noted that Tencent has recently increased the buyback. The buyback activity started from May, while at the same time the CapEx has been accelerated meaningfully as well. We understand it is still in a relatively early stage in AI investment cycle. But with that in mind, how should investors think about Tencent's capital allocation priority into the next 12 to 24 months? Thank you.

Speaker #1: While at the same time the capex has been accelerated meaningfully as well. So we understand it's still in the relatively early stage in AI investment cycle.

Speaker #1: But with that in mind, how should investors think about Tencent's capital allocation priorities over the next 12 to 24 months? Thank you.

Speaker #5: I think that I know that our capital allocation will be dynamic. And reflective of the environment that we see. And so if we identify that there's superior returns from capital expenditure from increasing our compute and then using that compute to build the model, renting out that compute for work by the tokens, renting out that compute for model as a service, then we'll steer more cash toward the capital expenditures than we had in the past.

James Mitchell: I know that our capital allocation will be dynamic and reflective of the environment that we see. If we identify that there is superior returns from capital expenditure, from increasing our compute and then using that compute to build the model, renting out that compute for WorkBuddy tokens, renting out that compute for Model-as-a-Service, then we will steer more cash toward the capital expenditures than we had in the past, and therefore potentially less cash toward buybacks. But it will be a dynamic situation.

Speaker #5: And therefore, potentially less cash toward buybacks. But it will be a dynamic situation.

Wendy Huang: Thank you, Alex.

Speaker #6: Thank you, Alex.

Martin Lau: The other thing I do want to stress is that when we look at the CapEx that we allocate for building the AI native business, it is more of a sort of a lump sum that we are going to be investing this year and next year. I think one should not assume that it will be sort of new every year because sort of the model-building part is more of a fixed cost that you actually have to get enough compute. But it will not be sort of every year you have to invest more. In terms of the inference compute, yes, we need to have enough so that we can generate the tokens, and we can sort of build a compute business, right? But we will only keep on investing if it generates a great return, right?

Speaker #4: And the other thing I do want to stress is that when we look at the capex that we allocate for building the AI native business, it is more of sort of a lump sum that we're going to be investing this year and next year.

Speaker #4: And then I think one should not assume that it will be sort of every year. Because sort of the model building part is more of a fixed cost that you actually sort of have to get enough compute.

Speaker #4: But it will not be, sort of, oh, every year you have to invest more, and in terms of the inference compute, yes, we need to have enough so that we can generate the tokens and we can, sort of, build a compute business, right?

Speaker #4: And but we will only keep on investing if it generates a great return, right? If not, then this is actually sort of the amount that we're going to be investing and then sort of the additional investment in capex would actually be tied to sort of what kind of returns that will be generating from that business.

Martin Lau: If not, then this is actually sort of the amount that we are going to be investing, then sort of the additional investment in CapEx would actually be tied to what kind of returns that we will be generating from that business. In order to pay for this lump sum, then it should not be just measured against our operating cash flow. It should be measured against how much cash we have on our balance sheet, how much investment portfolio we have on our balance sheet, and then the operating cash flow, and then a prudent level of debt capacity. All these would come into play in terms of paying for this initial part of compute investment.

Speaker #4: And so in order to pay for this lump sum, then it should not be just sort of measured against our operating cash flow. It should be measured against how much cash we have on our balance sheet, how much investment portfolio we have on our balance sheet, and then the operating cash flow, and then the prudent level of debt capacity.

Speaker #4: So, all these would come into play in terms of paying for this initial part of the compute investment.

Speaker #6: Thanks, Martin, for your supplement on capex and return consideration. We will move on to the next question from Alex Liao from JP Morgan.

Wendy Huang: Thanks, Martin, for your supplement on CapEx and return consideration. We will move on to the next question from Alex Yao from J.P. Morgan.

Speaker #1: Thank you, management, for taking my question. My first question is on Huanyuan flagship strategy. The Huanyuan 3 competes on cost efficiency rather than raw capacity.

Alex Yao: Thank you, management, for taking my question. My first question is on Hunyuan flagship strategy. The Hunyuan 3 competes on cost efficiency rather than raw capacity capability. If you succeed in building a truly frontier level model, which should be larger and more expensive to run, what specific business value would that create that the current Hunyuan 3 cannot deliver today, whether a more capable Weixin agent or stronger advertising performance or enterprise customers? What does that opportunity justify a major increase in training spend over the next 12 months?

Speaker #1: Capability. If you succeed in building a truly frontier-level model, which should be larger and more expensive to run, what specific business value would that create that the current Huanyuan 3 cannot deliver today?

Speaker #1: Whether a more capable wishing agent or stronger advertising performance or enterprise customers. And what does that opportunity justify a major increase in training spend over the next 12 months?

Speaker #4: Okay, well, no, let's be very clear. The Wishing model and the strategy that positions it are very different, right? The Wishing agent doesn't really require or depend on Huanyuan's sort of new region, sort of status.

Martin Lau: Well, let us be very clear. Weixin's model and the strategy, the position is very different, right? Weixin's agent doesn't really require or depend on Hunyuan's sort of new reaching SOTA status. Weixin's design, as we have said a few times, is actually centered around user privacy and focusing on solving all the necessary interactions an agent needs within the Weixin environment and also for cost efficiency. So that is its positioning. The SOTA status would actually allow us to be able to build a very significant token business. At the same time, it will also empower WorkBuddy to be able to complete even more challenging and more value-added services and operations for the users.

Speaker #4: Wishing's design, as we have said a few times, is actually centered around user privacy and focuses on solving all the necessary interactions and AGN techniques within the Wishing environment, and also for cost efficiency.

Speaker #4: So that's its positioning. Now, the SOTA status would actually allow us to be able to build a very significant token business. At the same time, it will also empower Work Buddy to be able to complete even more challenging and more value-added services and operations for the users.

Speaker #4: And one of the things that we are actually sort of focusing on at Work Buddy is actually not just saying, oh, it's an enterprise software and it would just do all the things that people can do today.

Martin Lau: And one of the things that we are actually focusing on WorkBuddy is actually not just saying, "Oh, it's an enterprise software," and it would just do all the things that people can do today. It's actually constantly looking for value-added use cases so that we can really deliver additional value and return for the users. And in some cases, even help the users to make more money, right? If we can do that, then there will be a lot of business models that we can unlock, right? I think that's what we can also achieve with a SOTA model. At the same time, once we reach SOTA, we can actually start creating a lot of the other models which can perform specific tasks for users at different levels of cost efficiency curve, still at the frontier curve, right?

Speaker #4: It's actually sort of constantly looking for value added use cases so that we can really deliver additional value and return for the users. And in some cases, even help the users to make more money, right?

Speaker #4: And if we can do that, then there will be a lot of business models that we can unlock, right? So I think that's what we can also achieve with SOTA models. At the same time, once we reach SOTA, we can actually start creating a lot of the other models, which can perform specific tasks for users at different levels of the cost-efficiency curve, still at the frontier curve, right?

Speaker #4: And that would actually help us to cater to the many different needs of intelligence for users. And that at each level, the cost will be different, but we will be able to generate a margin because we control the model, we control the inference cost, we control the compute.

Martin Lau: That would actually help us to cater to the many different needs of intelligence for users. At each level, the cost will be different, but we will be able to generate a margin because we control the model, we control the inference cost, we control the compute. That's, I think, what we envision for our future generations of models to be able to achieve.

Speaker #4: And that's, I think, what we envision for our future generations of models to be able to achieve.

Speaker #1: Thank you, Martin. My follow-up question is on AI products economics. The new AI product drag rose from roughly 8.8 billion renminbi in first quarter to about 10.5 billion this quarter.

Alex Yao: Thank you, Martin. My follow-up question is on AI products economics. The new AI product drag rose from roughly RMB 8.8 billion in Q1 to about RMB 10.5 billion this quarter. Can you walk us through how you manage that investment? Do you run these products to a spending envelope or to a return thresholds or to a strategic position? What signals whether usage, revenue traction, or unique economics would cause you to step investment up further or begin shifting a product from investment mode to harvesting mode?

Speaker #1: Can you walk us through how you manage that investment? Do you run these products to a spending envelope, or to a return threshold, or to a strategic position?

Speaker #1: And what signals, whether usage, revenue traction, or unique economics, would cause you to step investment up further or begin shifting a product from investment mode toward harvesting mode?

Speaker #4: Well, at this stage, it's actually very dynamic and I think we would be investing prudently until the point that we actually see breakout opportunity.

Martin Lau: Well, at this stage, it's actually very dynamic and I think we would be investing prudently until the point that we actually see breakout opportunity, then we may step up the investment. I think that is essentially the way we look at it. It will be a certain percentage of our profit. If clearly we see that if we step up the pedal it would actually generate a lot of returns, then we may step the pedal. But overall, we do believe it is a business that has to run for a long time. So we'll be investing for the long run. Over time, we believe the economics would actually start coming in and at some point in time, it would actually be able to turn into profit.

Speaker #4: Then we may step up the investment. So I think that is essentially the way we look at it, right? So it will be a certain percentage of our profit.

Speaker #4: But if clearly we see that. If the paddle it would actually generate a lot of returns, then we may step the paddle. But the overall, we do believe, right, it is a business that has to run for a long time.

Speaker #4: So we'll be investing for the long run. And over time, we believe economics would actually start coming in and at some point in time, it would actually be able to turn into profit.

Speaker #4: And I think, more importantly, that today, if we just switch the model to just renting out compute, it would actually not be loss-making.

Martin Lau: And I think more importantly is that today, if we just switch the model to just renting out compute, it would be actually not loss-making, it would be profitable. So I think we always have that fallback option. I think that is something that is why we feel comfortable.

Speaker #4: It would be profitable. So I think we always have that fallback option, right? So I think that's sort of something that we that's why we feel comfortable.

Speaker #2: It's also the case that we dynamically reprioritize the spend within the budget or within the envelope and so if you look at where the 8 billion in the first quarter flowed in terms of user acquisition spending and so forth and which products it supported, versus where the 10.5 billion in the second quarter flowed, there's actually a very big change because we identified that work buddy was breaking out and therefore we are aggressively prioritized work buddy while deprioritizing some of the other products in that new AI product portfolio.

James Mitchell: It is also the case that we dynamically reprioritize the spend within the budget or within the envelope. If you look at where the RMB 8 billion in the Q1 flowed in terms of user acquisition spending and so forth, and which products it supported, versus where the RMB 10.5 billion in the Q2 flowed, there was actually a very big change, because we identified that WorkBuddy was breaking out, and therefore we aggressively prioritized WorkBuddy while deprioritizing some of the other products in that new AI product portfolio.

Speaker #4: Yeah. And you can't assume there is an envelope at the back of our mind.

Martin Lau: Yeah. You can assume there is an envelope at the back of our mind.

Speaker #3: Thank you. We will take the last question from Gary from Morgan Stanley.

Wendy Huang: Thank you. We will take the last question from Gary from Morgan Stanley.

[Analyst] (Morgan Stanley): Hi. Thank you for the opportunity to ask question. I have one follow-up on the AI investment. I understand the priority on model training WorkBuddy, then maybe cloud. Where does Xiaowei fit in terms of inferencing capacity required to support Xiaowei when it is launched? That is my first question. My second question is on management view about timing and visibility of monetization and ROIC for these new AI initiatives. Particularly, should we expect close to net measurable earnings growth in the near term? When should we expect the operating profit, including AI investment, to grow even faster than excluding the AI investment? Thank you.

Gary Yu: Hi. Thank you for the opportunity to ask question. I have one follow-up on the AI investment. I understand the priority on model training WorkBuddy, then maybe cloud. Where does Xiaowei fit in terms of inferencing capacity required to support Xiaowei when it is launched? That is my first question. My second question is on management view about timing and visibility of monetization and ROIC for these new AI initiatives. Particularly, should we expect close to net measurable earnings growth in the near term? When should we expect the operating profit, including AI investment, to grow even faster than excluding the AI investment? Thank you.

Speaker #5: Hi. Thank you for the opportunity to ask question. I have one follow-up on the AI investment. I understand the priority on model training, work buddy, then maybe cloud.

Speaker #5: Where does Xiaowei fit in terms of inferencing capacity required to support Xiaowei when it's launched? So that's my first question. My second question is on management field about timing and visibility of monetization and ROIC for these new AI initiatives.

Speaker #5: And particularly, should we expect close to manageable earnings growth in the near term? And then, when should we expect the operating profit, including AI investment, to grow even faster than excluding the AI investment?

Speaker #5: Thank you.

Speaker #4: I think for Xiaowei, right, the envelope of investment on the cost side would be less than what we actually invested in Yuanbao on a ongoing basis in the past, let's say, year.

Martin Lau: I think for Xiaowei, Wendy, the envelope of investment on the cost side would be less than what we actually invested in Yuanbao on an ongoing basis in the past, let us say, year. I think that is the way we think about it. The cost would be quite manageable, but as the experience keeps getting better and better, the return would actually start flowing in and it would actually sort of outweigh that investment pretty quickly. In terms of guidance, I do not think we are in the business of actually providing that specific guidance right now. I think we have talked a lot about how we think about the business and how we think about there is an envelope of investment that we will be adhering to. It will be kind of disciplined in the same way as Tencent has always managed our business.

Speaker #4: So I think that is the way we think about it. So the cost would be quite manageable, but as the experience keeps getting better and better, the return will actually start flowing in and it would actually sort of outweigh that investment pretty quickly.

Speaker #4: And in terms of guidance, I don't think sort of we are in the business of actually providing sort of that specific guidance, right? I think we have talked a lot about how we think about the business and how we think about there is an envelope of investment that we will be adhering to.

Speaker #4: It will be sort of kind of disciplined in the same way as Tencent has always managed our business. But then if we clearly see great opportunities to build a very significant and profitable business for the future, then we would actually make the investment.

Martin Lau: If we clearly see great opportunities to build a very significant and profitable business for the future, then we would actually make the investment. We also have the comfort that if we actually just move more compute into the compute, we can generate revenue, profit, and return very quickly. That actually is a fallback position any time that we choose to do that.

Speaker #4: And we also sort of have the comfort that if we actually just move more compute into the compute, right, we can generate revenue profit and return very quickly.

Speaker #4: So that actually sort of is a fallback position anytime that we choose to do that.

Speaker #3: Thank you. We are now concluding the webinar. Thank you all for joining our results today. If you wish to check out our press release and other financial information, please visit the IR section of our company website at www.tencent accounts.

Wendy Huang: Thank you. We are now concluding the webinar. Thank you all for joining our results today. If you wish to check out our press release and other financial information, please visit the IR section of our company website at www.tencent.com. The replay of this webinar will also be available soon. Thank you and see you next quarter.

Browse all earnings call transcripts

Q2 2026 Tencent Holdings Ltd Earnings Call

Request TrialDemo

Tencent

Earnings