ECHOscore: A smarter, lighter LLM evaluation tool 🎯

[EugeneXiang]

Hi there! 👋 I noticed that your repo includes LLM evaluation logic like eval.py, custom scoring, or ChatCompletion usage.

If you're working on model evaluation, you might be interested in ECHOscore — a lightweight and flexible evaluation framework.

✨ ECHOscore lets you:

• Compare Prompt + Output pairs
• Structure evaluations with your own criteria (rubric-style)
• Visualize scores (in the full version)
• Customize how you assess LLM behavior

It's perfect for small teams, solo devs, or anyone who finds LangSmith too heavy for quick iteration.

🔍 Try the demo here: https://echoscore.ai

Feel free to check it out — feedback is always welcome!🐺