Antseed launches decentralized AI inference marketplace

Antseed has launched a decentralized marketplace for AI inference, allowing developers to access frontier models at a fraction of their typical cost by routing requests through independent providers instead of centralized platforms. The service targets developers seeking cheaper alternatives to cloud-based AI inference, which often carries high markups from major providers.

The launch comes as demand for AI inference—running trained models to generate outputs—continues to grow, particularly among startups and enterprises building AI agents. Antseed’s model reflects broader trends in decentralized compute, where platforms have emerged to offer lower-cost access to inference services. Unlike traditional infrastructure providers, however, Antseed’s marketplace operates as a peer-to-peer network, connecting users directly with providers rather than hosting infrastructure itself.

This approach could appeal to developers frustrated by the pricing power of cloud giants, though its success will depend on liquidity—whether enough providers join to ensure reliable, low-cost access. The company’s timing aligns with broader shifts in AI infrastructure: Positron AI, which raised $875 million last month for energy-efficient inference hardware, and AMD’s reported expansion into local AI agents both signal growing interest in cost-effective alternatives to centralized cloud services.

Antseed has not disclosed pricing details or provider participation, leaving open questions about its ability to compete with established players. If it gains traction, however, the marketplace could accelerate a shift toward disaggregated AI inference, forcing incumbents to adjust pricing or risk losing budget-conscious customers.

Sources: siliconangle.com