Hugging Face Cloudflare Workers AI Inference Integration Discontinued

AI Impact Summary

The Hugging Face integration with Cloudflare Workers AI for serverless GPU inference has been discontinued. Developers should migrate to the Hugging Face Inference API, Inference Endpoints, or other deployment options. This change impacts workflows relying on the previously available integration, particularly those utilizing models like Llama 2 7B or Mistral 7B for RAG applications, which could have incurred daily inference costs of around $1.