Mininglamp Technology (2718.HK) Invited to the Inaugural Beijing-Birmingham Technology Week to Accelerate Overseas Market Expansion

It is understood that Birmingham Tech Week is the largest regional official tech brand event in the UK, gathering high-quality European innovative enterprises, industrial capital, and professional incubation vehicles. This event marks the brand's first landing in China.

Mininglamp Technology (2718.HK) was one of the four Beijing sci-tech enterprises invited to participate in the event. At the event site, alongside other invited enterprises, Mininglamp Technology promoted its self-developed innovative achievements, shared practical experiences in international cooperation and overseas market expansion, and expressed its willingness to enter the UK market and deepen industrial cooperation. The event also unveiled the "China-UK Enterprise Mutual Visit Program," granting Mininglamp Technology the opportunity to participate in the main event of Birmingham Tech Week this October.

According to public records, Mininglamp Technology had previously signed an agreement to become a partner enterprise in the fifth batch of strategic enterprises of the Hong Kong SAR Government, establishing its international headquarters at Cyberport.

As a large model concept stock, Mininglamp Technology is dubbed by the market as the "First Agentic AI Stock in Hong Kong Equities." In addition to its self-developed DeepMiner large model product line (underlying engine), model groups including Mano/Cito, and the Cider inference acceleration framework, the company has also built the Octo platform layer to connect human-Agent collaborations into a centralized hub. Delivered ultimately in the form of Agentic Services, it empowers the implementation of decision-making AI agents across industries such as marketing and mass consumption.

Since 2026, Mininglamp Technology has consecutively deployed three key strategic moves in the direction of Agentic AI.

First, open-sourcing edge-side models and inference frameworks. The company officially open-sourced Mano-P, an edge-side GUI-VLA agent model that supports running locally on Mac to comprehend and operate graphical interfaces via a vision-only approach, eliminating reliance on API integrations. The 72B version of this model ranked first globally on the OSWorld specialized model leaderboard with a success rate of 58.2%. The concurrently open-sourced Cider inference acceleration SDK boosts operator speeds by 1.4 to 1.9 times under the W8A8 mode compared to native MLX.

Second, launching the AI-Native hardware Octic. On May 13, Mininglamp Technology introduced its first AI recording hardware tailored for meeting scenarios. Designed in a magnetic card form factor, it is positioned to "accompany you in meetings." Its core innovation lies in "Private AI"—where data is processed locally on the device, understanding the user better the more it is used. With this, Mininglamp Technology officially completed a hardware-software integrated closed loop from perception (Octic) to processing (Mano-P) and collaboration (Octo).

Third, the Octo platform transforming Agents into organization-level digital workforces. Octo is the world's first open-source and trusted Agent collaboration network. It has been internally deployed among roughly 1,400 employees, with over 2,900 AI Agents running efficiently in real-world workflows. The company treats Agents as organizational members of equal status to humans, driving AI Agents to transition from isolated personal efficiency assistants toward a collaborative, orchestratable, and evaluable organization-level digital workforce.

These three moves have completely pieced together Mininglamp Technology's product matrix into a three-tier architecture of "Model—Platform—Hardware." Mininglamp Technology has become the only AI company in the Hong Kong stock market to concurrently possess a four-tier capability spanning edge-side models, inference engines, Agent collaboration platforms, and hardware entry points.

In early June 2026, Mininglamp Technology announced the signing of a memorandum of strategic cooperation with The Education University of Hong Kong to jointly build an "AI + Education" joint research platform, opening up AI tools such as DeepMiner, Octo, and CoCraft to extend edge-side AI capabilities into the fields of education and scientific research. Following this, it signed a strategic cooperation framework agreement with PATEO Connect+, a Hong Kong-listed company, extending its Agentic AI strategy from marketing, office work, and content production scenarios into the trillion-RMB track of intelligent connected vehicles.

CICC pointed out in its research report that it is bullish on the long-term development potential of the company's Agentic Services business. Its business model has upgraded from traditional "software selling" to outcome-based charging, substituting intensive human labor with computing power drivers, thereby deeply binding with customer delivery outcomes. Current application scenarios include the closed loop of performance advertising, brand advertising, content e-commerce, and micro-drama production and placement. The company believes that edge-side models hold advantages in data security and marginal costs. Looking ahead to 2026, management stated that Agentic Services are expected to extend into segments beyond marketing and advance globalization, becoming an important source of revenue contribution.

Risk Warning: The above content is an objective compilation of public information for reference purposes only and does not constitute investment advice. The AI industry is subject to multiple factors such as technological iterations and market competition, and uncertainties exist. Investing involves risks; caution is required when entering the market.

NewTimeSpace Disclaimer: All content herein is the original work of NewTimeSpace. Any reproduction, reprinting, or use of this content in any other manner must clearly indicate the source as "NewTimeSpace". NewTimeSpace and its authorized third-party information providers strive to ensure the accuracy and reliability of the data, but do not guarantee the absolute correctness thereof. This content is for reference only and does not constitute any investment advice. All transaction risks shall be borne by the user.