Embodied intelligence data infrastructure company SCALEFORCE has recently closed a funding round worth tens of millions of yuan. Proceeds will go toward iterating its physical AI operating system MatrixOS, building a large-scale data production network, and expanding the team. The founding team hails from Baidu, Meta, and Huawei, and the company secured a multi-million-unit data order from Tuoshi Zhihang within its first three months.
Funding Overview: Tens of Millions Enter the Embodied Intelligence Data Race
SCALEFORCE, an embodied intelligence data infrastructure company, has announced a funding round of tens of millions of yuan. The company stated that proceeds will be deployed across three strategic priorities:
- Technical iteration of the physical AI operating system MatrixOS
- Construction of a large-scale data production network
- Team expansion
These moves aim to accelerate SCALEFORCE's rise as a global leader in embodied intelligence data infrastructure.
Core Product: Three Technical Pillars of MatrixOS
Founded in May 2026, SCALEFORCE's core team comes from Baidu, Meta, and Huawei. The company has built an end-to-end Data/AI Infra platform, MatrixOS, which achieves a complete data flywheel loop across "data—algorithms—scenarios."
MatrixOS delivers three key advantages:
Strong Generalization
Powered by the ADA data generalization engine, MatrixOS uses the Latent Action Representation method and the JEPA unsupervised training paradigm. By jointly modeling training and evaluation data and mapping them into an action-tactile semantic latent space, the system decouples data from hardware configurations. This allows the same data assets to be efficiently reused across different robot bodies.
High Efficiency
The proprietary GDP (Global-Deep-Proactive) data quality engine leverages information entropy analysis, posterior probability estimation, action semantic analysis, and Bayesian active learning. It continuously optimizes collection strategies from a global perspective, achieving a data usability conversion rate of 80%.
Massive Scalability
SCALEFORCE has built a global, scaled physical AI data production network. The platform supports unified management and scheduling of hundreds of thousands of data collection nodes worldwide. It achieves end-to-end automation exceeding 95%. Furthermore, the company was the first globally to launch a sub-millisecond, multi-sensor time synchronization kit, enabling synchronized, large-scale collection of full-modal data including visual, tactile, force, and behavioral inputs.
| MatrixOS Capability | Technical Approach | Key Metric |
|---|---|---|
| Data Generalization | ADA engine + JEPA paradigm | Cross-robot body reuse |
| Data Quality | GDP quality engine | 80% conversion rate |
| Scalability | Global node network | 95%+ end-to-end automation |
Commercial Traction: Multi-Million Order from Tuoshi Zhihang Underway
MatrixOS's data flywheel can operate independently or be integrated with clients to build joint big-data loops. SCALEFORCE has established deep partnerships with numerous world model companies, leading embodied intelligence robot manufacturers, and industry solution providers, including Tuoshi Zhihang and Zhizai WuJie.
A noteworthy commercial case: SCALEFORCE's multi-million-unit data order from Tuoshi Zhihang is now fully underway. Through its "targeted scenario data collection + MatrixOS platform data processing" service model, SCALEFORCE helped Tuoshi Zhihang's A-series robots successfully enter Aptiv factories.
Additionally, SCALEFORCE has joined the Huawei Ascend computing ecosystem and Xinhua Net's "Robot+" real-world training scenario co-construction plan. As a first-batch member, the company is deeply involved in scenario implementation and data standard development. It is also actively building the "Spark" industry ecosystem partnership program.
International Expansion: Southeast Asia Team Assembled
Notably, SCALEFORCE has substantively launched its international expansion. An overseas team led by senior Southeast Asian experts has been assembled, aiming to systematically advance overseas business and partner networks. This signals that global competition in embodied intelligence data infrastructure is already underway.
Industry Perspective: Why Embodied Intelligence Data Is the New Frontier
Embodied intelligence is widely regarded as the next major frontier in AI development. Unlike pure software AI, embodied intelligence requires robots to perform tasks in real physical environments, creating exponentially growing demand for high-quality, multi-modal, generalizable training data. SCALEFORCE targets precisely this critical layer — data infrastructure.
Traditional data annotation models struggle to meet embodied intelligence's granular requirements for action-level, tactile-level, and force-level data. MatrixOS addresses the industry's persistent "data bottleneck" through architectural innovation at the operating-system level, forming a closed loop of data collection, processing, generalization, and reuse. For more information, visit EX1000.COM.
