Three new features now available in Public Preview enable you to utilize vision-based models in the model catalog, enjoy seamless integration with LLMOps through the Prompt Flow SDK/CLI and VS Code extensions, and monitor LLM completions for safety and quality metrics.
Vision models for model catalog: You can now utilize new task types, namely image classification, image segmentation, and object detection, to finetune, evaluate, and deploy base models using your own training data.
Enjoy the Prompt Flow code-first functionality: You can now fully leverage the power of LLMs, start quickly with pre-built workflows and evaluation metrics, and enjoy high-efficiency iterations with LLMOps integration into Prompt Flow.
Model Monitoring for LLM Applications: You can now monitor LLM applications in production for safety and quality on a user-defined cadence to ensure they deliver maximum business impact.
