The company is launching Workers AI with a curated set of popular, open-source models that cover a wide range of inference tasks. It is also partnering with Hugging Face, a leading AI community and hub, to allow developers to browse and run a subset of the Hugging Face catalog directly in Workers AI. The company plans to expand the number of models based on community feedback. It is also working on analytics and observability capabilities to provide insights into usage, performance, and spend on a per-model basis.
Key takeaways:
- Cloudflare has launched Workers AI, a service that allows developers to run AI models with just a few lines of code, powered by a global network of GPUs.
- The service is designed to be open, accessible, serverless, privacy-focused, and pay-as-you-go, with a focus on a best-in-class developer experience.
- Workers AI is launching with a curated set of popular, open-source models, and plans to expand based on community feedback. A partnership with Hugging Face will soon allow developers to browse and run a subset of the Hugging Face catalog directly in Workers AI.
- Cloudflare is also launching Vectorize, a vector database designed to work seamlessly with Workers AI, and an AI Gateway for running APIs with features like caching, rate-limiting, analytics, and logging.