Mozilla Report: Open-Source AI Wins Tokens, Lags in Production
Mozilla's 2026 report shows open-source AI models leading in token volume but trailing closed models in production deployment.
Mozilla's 2026 State of Open Source AI report finds open-weight models now handle a majority of tokens routed through OpenRouter, with the five highest-volume models all open. The capability gap versus closed frontier models has narrowed from roughly 8% to about 3%, now concentrated in reasoning and long-context tasks, while GPT-4-class inference costs fell around 50x in three years.
According to a Mozilla/SlashData survey of 1,410 developers, open models lead in adoption (79% versus 71% for closed models) but lag in shipping to production: only 51% of teams using open models reach production, versus 63% for closed. Enterprise scale closes that gap for closed models but barely moves the needle for open ones, pointing to an operational and trust problem rather than a capability one.
Mozilla's 48-component stack analysis shows open-source AI scoring strong on core capability but consistently weak on standardization and enterprise readiness across every layer — the operational gap engineers actually hit when deploying open models at scale.