Definition ∞ Inference services refer to the provision of artificial intelligence models for making predictions or decisions based on new, unseen data. These services take a pre-trained AI model and apply it to incoming information to generate outputs. They are a core component of deploying AI capabilities in real-world applications.
Context ∞ The emergence of decentralized inference services is a notable trend in Web3, allowing users to access AI models without relying on centralized cloud providers. This democratizes access to AI capabilities and promotes censorship resistance for AI applications. News often highlights protocols that enable distributed inference, potentially reducing costs and increasing transparency for AI-driven decision-making in digital asset markets.