AI marketing technology company Taitong Technology announced on August 20, 2026, that it has completed a new funding round led by Huatai Pan-Atlantic Fund, with participation from Kunpeng Guangyuan, GAC Capital, and GSR Ventures. The deal marks the first RMB-denominated investment project for Huatai Pan-Atlantic Fund, according to Sohu’s coverage of the announcement.
Taitong said the fresh capital will be directed toward deepening its proprietary Taiji Professional Large Model and Navos Marketing Multi-Agents, as well as recruiting top AI talent. The company is also expanding its layout in Agentic Commerce, Agent-to-Agent collaboration, and Generative Engine Optimization. The exact size of the round was not disclosed, Sina Finance reported.
Deepening Taiji and Navos for Enterprise Workflows
The funding focus centers on two core assets that define Taitong’s technology roadmap. The Taiji Professional Large Model underpins a range of marketing and reasoning capabilities, and the company’s Taiji Q&A reasoning model has ranked first globally in the SuperCLUE advertising and marketing large model evaluation. That recognition situates Taiji as a flagship component of the portfolio and as an engine for tasks that demand both domain grounding and decision support across campaign and content workflows.
Navos, Taitong’s enterprise AI workforce, moved to version 2.0 in July 2026. The company positions Navos 2.0 as an all-in-one, enterprise-level AI workforce, integrating a suite of marketing agents designed to coordinate tasks end to end. Framed alongside Taiji’s reasoning strengths, the Navos Marketing Multi-Agents concept points to a coordinated agent mesh that can plan, execute, and refine activities across creative production, audience targeting, and performance feedback loops. Taitong’s emphasis on multi-agent orchestration suggests a strategy that treats marketing operations as a living system, where specialized agents act in concert rather than as isolated tools.
Together, Taiji and Navos form a stack that advances beyond basic content generation. The goal is to enable teams to delegate complex, multi-step marketing tasks to a digital workforce that can reason, coordinate, and adapt. The latest funding aims to accelerate this integration, while investment in top-tier AI talent signals that Taitong is preparing to scale model research, product engineering, and real-world deployment support.
Building Toward Agentic Commerce and A2A Collaboration
Beyond core models and workforce agents, Taitong is expanding in three related arenas: Agentic Commerce, Agent-to-Agent collaboration, and Generative Engine Optimization. Agentic Commerce envisions autonomous or semi-autonomous agents that help discover products, design campaigns, and transact within defined guardrails. A2A collaboration extends that vision by synchronizing agents that specialize in distinct tasks, reducing handoff friction between planning, buying, and measurement. GEO focuses on how generative models interact with discovery systems, steering content so that algorithms can parse and rank it more effectively within policy and quality frameworks.
These vectors together outline a path in which marketing agents evolve from tools to coordinated actors within commercial ecosystems. As companies look for leverage in how content is created, distributed, and optimized, the ability for agents to collaborate directly becomes a core capability. In China, the interplay between agents, storefronts, and platforms is accelerating experimentation in digital trade, as discussed in our previous report on China’s AI agents and commerce.
The company’s decision to allocate proceeds to these initiatives indicates an intent to marry reasoning with execution. Taiji’s global performance in the SuperCLUE advertising and marketing evaluation provides an anchor for higher-order planning, while Navos 2.0 presents a container in which specialized marketing agents can cooperate. A2A collaboration then becomes the connective tissue that lets those agents negotiate tasks, exchange intermediate outputs, and converge on measurable outcomes without manual stitching.
GEO, meanwhile, speaks to the visibility problem facing any enterprise-grade content pipeline. If generative outputs are not interpretable by ranking and recommendation systems, even high-quality work can underperform. Focusing on GEO within the same investment cycle implies that Taitong wants to ensure its agent workforce is not only productive but also discoverable and measurable inside algorithmic environments.
Global Footprint and Strategic Partnerships
Taitong reports that it serves more than 100,000 enterprises operating across over 200 countries and regions. The company also partners with Meta, Google, TikTok, and OpenAI. That footprint and partner set frame Navos and Taiji as technologies built to operate across diverse ad networks, content surfaces, and developer ecosystems. For enterprises that need to manage campaigns in multiple markets, the ability to deploy, tune, and govern agents in parallel becomes a practical differentiator.
Partnerships across global platforms matter because marketing agents must respect policy, privacy, and format constraints that differ by channel. The presence of integrations or co-development relationships can help translate high-level strategies into channel-ready assets, pacing, and measurement. When agents know how to package creative and attribution for each surface, they reduce the operational burden on human teams and improve the fidelity of feedback signals that inform model updates.
The new funding, led by Huatai Pan-Atlantic Fund, also carries financial significance. As Huatai Pan-Atlantic Fund’s first RMB-denominated investment project, the deal indicates a willingness to back AI assets within a local currency structure. That choice can shape how the company aligns future financing, vendor contracts, and hiring, particularly as it seeks to recruit top AI talent. A RMB framework can also be compatible with domestic procurement and research timelines, which may be relevant as Taitong deepens model training and agent orchestration.
Recruiting remains a central lever in this plan. Scaling a professional model family and a multi-agent workforce typically requires teams that span model research, data engineering, prompt and policy design, tooling, and customer deployment. By signaling that a portion of proceeds is earmarked for talent, Taitong sets expectations that it will invest in the human capital needed to bring Taiji and Navos further into enterprise workflows.
The company’s roadmap suggests a move from point solutions to a durable operating system for marketing. With Taiji as a reasoning core, Navos as an orchestrated workforce, and expansions in Agentic Commerce, A2A collaboration, and GEO, the emphasis now shifts to reliability and governance. Enterprises will expect transparent controls, repeatability, and auditability from agents that touch budgets and brand assets. The SuperCLUE ranking gives Taiji a benchmark, and Navos 2.0 establishes a packaging layer that can be deployed across teams and regions. The newly completed funding round is intended to advance these priorities while leaving the headline size undisclosed, keeping the focus on technology depth and execution rather than capital scale.