Hugging Face has updated its inference endpoints with new analytics features, allowing users to track performance metrics such as latency, throughput, and model usage. These updates aim to provide developers with more insights into their models' behavior and help optimize them for better results. --- Why it matters: These updates matter because they enable engineers to fine-tune their models and improve their overall performance in real-world applications. Source: https://huggingface.co/blog/endpoint-analytics