Gartner predicts open-weight models will run 90% of enterprise agent workflow steps by 2030

Gartner forecasts that open-weight AI models will execute 90% of enterprise agent workflow steps by 2030, a shift the research firm says could significantly cut costs and give companies greater control over AI processes. The prediction, reported by Crypto Briefing, signals a potential turning point in enterprise AI adoption, as open-weight alternatives to proprietary models from vendors like OpenAI and Anthropic gain traction. The move matters for businesses weighing the trade-offs between the flexibility and cost savings of self-hosted models against the convenience of closed, API-based systems. Investors and enterprise buyers should watch how quickly open-weight model performance closes the gap with proprietary systems, and whether major cloud providers accelerate support for open-weight deployments.