Official Launch | 2026 China AI Progress and Breakthrough Annual Best Application Practice Selection Campaign

With the continuous development of artificial intelligence technology and applications, the evaluation focus of AI Agents has gradually extended to task execution, business collaboration, and actual application outcomes. For different product forms such as enterprise general tasks, personal services, smart terminals, and the physical world, the industry needs to combine real-world scenarios to assess their capability boundaries, implementation value, and continuous improvement performance. At the same time, the advancement of technological innovation and industrial applications also depends on the continuous contributions of researchers, entrepreneurs, and industry promoters. Systematically reviewing representative figures and application practices will help present the innovative achievements and development direction of China's AI industry.

Frost & Sullivan, in collaboration with LeadLeo Research Institute, has launched the "2026 China AI Progress and Leap Annual Best Application Practice" research and selection campaign. Eligible enterprises, institutions, and individuals are welcome to submit applications. Through documentation submission, research surveys, multi-dimensional evaluation, and expert review, we will identify representative AI heroes and annual benchmark intelligent agents, providing references for technological innovation, industrial applications, and industry research.

"2026 China AI Progress and Leap Annual Best Application Practice" focuses on enterprises and representative products in the field of intelligent agents in China, covering eight major areas: enterprise-level general, consumer-level, edge AI, physical AI, enterprise services, industrial implementation, innovative breakthroughs, and social value. Additionally, the research specifically sets up a "AI Heroes" selection category, highlighting industry leaders and promoters who have made contributions to technological innovation, industrial implementation, business model exploration, and ecosystem construction, to demonstrate the role of individual efforts in the AI industrialization process.

This research will systematically present the reasons for selection of AI heroes and annual benchmark intelligent agents, their representative achievements, and application outcomes, as well as the practical progress in areas such as intelligent capability advancement, intelligent form leap, application value deepening, and annual special benchmarks, promoting the exchange of innovative achievements and industrial ecosystem collaboration.

Frost & Sullivan, in collaboration with LeadLeo Research Institute, is now inviting relevant enterprises, institutions, and individuals to participate. We welcome the submission of verifiable technical achievements, product information, and application practice materials to jointly present the annual innovative and application achievements of China's AI industry.

1. In-depth Interpretation of the Annual Best Application Practice

1. What representative characteristics do AI heroes exhibit in verifiable contributions, industrial effectiveness, innovation leadership, and value and impact?

2. What trends in the development of AI talent, technological innovation, and industrial collaboration are reflected in the achievements and practices of representative figures?

3. What evolution directions in capabilities are shown by the eight types of benchmark intelligent agents in task execution, interaction methods, deployment forms, and system collaboration?

4. How are application outcomes verified in different scenarios? Which practices have the potential for large-scale application, industry replication, and continuous value creation?

2. Scan of High-Quality Targets and Figures

This research will conduct comprehensive studies on AI Agent application products and AI heroes, identifying benchmark products and representative figures with outstanding performance in technological innovation, market potential, and application value.

Enterprises, institutions, and individuals participating in this report collection will have the following opportunities:

1. By presenting verifiable technical contributions, product capabilities, and application practices through research results, you can increase the industry's understanding of relevant achievements;

2. Combining categorized evaluations and research feedback, we will analyze the scenario adaptation ability, application outcomes, and future improvement directions of products, enhancing potential customers and industrial partners' understanding of relevant products and application practices;

3. Using the research platform, you can establish connections with leading enterprises in other industries and explore cross-industry cooperation opportunities;

4. You will have the opportunity to be invited to participate in LeadLeo online/offline industry report insights meetings, and communicate with industry experts, industrial institutions, and investment institutions.

This selection consists of AI heroes and the eight types of benchmark intelligent agents. The eight types of benchmark intelligent agents are structured in "four groups of directions, eight types of selections" according to the "four dimensions: AI intelligent capability advancement, AI intelligent form leap, AI application value deepening, and annual special benchmark." The selection results are based on comprehensive research and evaluation in each direction. There is no order of selection for the selected items. The specific directions are as follows:

1.AI Intelligent Capability Advancement:Enterprise-level general benchmark Agent, targeting general enterprise tasks, focusing on multi-task processing, tool invocation, and cross-system execution; consumer-level benchmark Agent, targeting individual users, focusing on goal understanding, personalized interaction, and personal task execution.

2.AI Intelligent Form Leap:Edge AI benchmark Agent, targeting AI phones, AI PCs, and other smart terminals, focusing on edge reasoning, device capability invocation, and end-cloud collaboration; physical AI benchmark Agent, targeting perception, decision-making, and task execution in the physical environment, focusing on environment understanding, intelligent decision-making, and actual application capabilities.

3.AI Application Value Deepening:Enterprise service benchmark Agent, targeting enterprise service scenarios such as customer service, marketing, human resources, finance and taxation, and supply chain, focusing on professional service capabilities, system collaboration, and business outcomes; industrial implementation benchmark Agent, targeting industry applications such as finance, healthcare, industry, and government, focusing on industry scenario adaptation, actual application outcomes, and large-scale deployment capabilities.

4.Annual Special Benchmark:Annual innovative breakthrough benchmark Agent, focusing on original technology, key capabilities, and application form innovation, focusing on technological breakthroughs, application outcomes, and conversion potential; annual social value benchmark Agent, targeting public welfare, education, elderly care assistance, and public service fields, focusing on service accessibility, actual benefit outcomes, and continuous service capabilities.

AI HeroesUsing the C-E-I-V evaluation model, a comprehensive evaluation is conducted from four dimensions: verifiable contribution, industrial effectiveness, innovation leadership, and value and impact. Verifiable contribution focuses on technical originality, intellectual property rights, open-source contributions, and standardization achievements; industrial effectiveness focuses on the large-scale application of technology and economic value transformation; innovation leadership focuses on innovative projects, achievement incubation, and cooperation practices; value and impact focus on long-term contributions in industry influence, social responsibility, and international cooperation. The application materials should explain the actual role of the candidates and provide corresponding achievements or practice evidence.

Annual Benchmark Intelligent AgentsA comprehensive evaluation is conducted from four dimensions: Product, Effectiveness, Business, and Competition, with differentiated indicators set according to the eight directions. The product dimension focuses on functional innovation, user value creation, and the enhancement of customer experience and application depth through intelligence; the effectiveness dimension focuses on quantifiable application outcomes, including cost optimization, efficiency improvement, and operational value; the business dimension focuses on business expansion, value realization, sustainable growth, and industrial chain extension capabilities; the competition dimension focuses on differential advantages, market response capabilities, and market position and resilience in peer competition.

This selection will be conducted by a review team composed of Frost & Sullivan, LeadLeo Research Institute, and experts from professional investment institutions, universities, scientific research institutions, etc. Categorized evaluations will be carried out based on different professional backgrounds.

1. Using a selection method that combines active screening by the organizer and voluntary submission of materials by enterprises, we aim to expand the coverage of the selection, increase participation, and make the final selection results more reasonable and referable.

2. After the application materials are entered, the research team will conduct material review, research, and comprehensive evaluation according to different selection directions, and determine the selected items in combination with expert opinions. If the same product meets the conditions of multiple directions, it can submit to multiple directions simultaneously. The selection results do not represent any qualification recognition or commercial recommendation.

Project Launch: October 2026

First Round Submission Deadline: December 1, 2026

Second Round Material Return Deadline: December 31, 2026

Report Release (Estimated): January 2027

Report Release Channels: 14 mainstream report release platforms such as Wind and Choice, supplemented by LeadLeo's self-media matrix (LeadLeo LeadLeo, WeChat, Weibo, Toutiao, Xueqiu, Futu Niuniu, Shenhua, etc.) to reach more than 5,000 high-quality media, investment institutions, and other resources, for industry enterprises, research institutions, and related market participants to refer to.

This collection starts immediately. The first round of submissions closes on December 1, 2026. We sincerely invite eligible enterprises, institutions, and individuals to submit basic application information and relevant achievement materials via the QR code below. The research team will conduct preliminary screening based on the application situation and send the second round of material submission emails to the applicants who enter the next stage. The second round of material return deadline is December 31, 2026.

Long-press to recognize the QR code to participate in the collection