Natix Network and Valeo announced a partnership to combine Natix’s Solana-based DePIN data streams with Valeo’s automotive world-modeling capabilities. The aim is an open-source, multi-camera World Foundation Model (WFM) built from decentralized, incentivized video and geospatial data—an asset class that could change how autonomous systems are trained.
The announcement by Natix Network described a joint effort to build what the firms termed one of the largest open-source multi-camera WFMs. WFMs extend foundation-model design into the physical domain: they aim to understand and predict four-dimensional (space and time) dynamics rather than just generate text.
Natix contributes decentralized data acquisition via a Solana-based DePIN. Its VX360 system captures multi-camera 360° footage from participating vehicles, while the Drive smartphone app crowdsources real-time camera and geospatial inputs.
The network’s scale was highlighted in the announcement and supporting materials: over 500.000 hours of video data, roughly 265.000 drivers and more than 220.000.000 km covered, figures the partners say underpin the dataset’s diversity—particularly valuable for edge-case coverage.
The deal pairs Natix’s VX360 and Drive app data pipeline with Valeo’s compute and perception expertise, creating scale in rare driving edge cases that centralized fleets often miss—an operational edge for model generalization and safety testing.
Why this matters for DePIN, AI and the auto supply chain
Valeo brings vehicle-level domain knowledge, perception stacks and compute resources, complementing Natix’s decentralized data. The pairing aims to accelerate model training cycles and encourage reproducibility through open-source release of models, datasets and tooling—an explicit strategy to broaden academic and industry contributions to physical AI research.
Using Solana for data ownership and incentives, the partnership reframes data acquisition economics: contributors are compensated and data provenance is recorded on-chain, which the announcement argued can reduce bottlenecks in centralized collection and improve dataset transparency. For developers, that can translate to richer training corpora and clearer audit trails for model behavior.
For traders and crypto infrastructure observers, the tie-up is a concrete use case for DePIN economics on Solana: it demonstrates a demand-side application where on-chain incentives and provenance intersect with high-value, physical-world AI training data. That could influence developer activity, token utility narratives and attention to Solana-based data networks.
Investors and engineering teams will now watch model releases and dataset governance rules as the first practical tests. The published WFM, the partners’ open-source cadence and any metrics tied to contributor rewards will function as the immediate proofs of viability—and the first market signals for how DePIN-driven datasets scale into commercial autonomous systems.
