A modular example demonstrating GEPA (Generative Evolutionary Prompt Adaptation) for prompt optimization using direct OpenAI API calls and MLflow integration. Status: Fully functional! This project ...
Production-ready model serving backend using FastAPI and MLflow, with Gradio UI. Why FastAPI instead of Ray Serve? Railway has a PIDs (process IDs) limit that prevents Ray Serve from running properly.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback