Worker AI

Run AI models (Llama, Stable Diffusion, Mistral) on serverless GPUs at the network edge with Cloudflare Workers AI.

Workers AI is Cloudflare's platform for running AI inference directly at the network edge: no GPU clusters to manage, just fast, scalable execution. This service deploys popular open-source models (like Llama 2 and Stable Diffusion) on serverless GPUs across Cloudflare’s global network, ensuring ultra-low latency for real-time applications. Developers utilize the Workers AI API or Workers/Pages integration to build GenAI applications, seamlessly connecting to related Cloudflare services (Vectorize for vector storage, R2 for data lake, AI Gateway for centralized monitoring). It eliminates infrastructure complexity, allowing you to focus on application logic while the platform handles deployment, optimization, and scaling.

https://www.cloudflare.com/products/workers-ai/