پلت فرم اپن سورس ارزیابی نرم افزارهای هوش مصنوعی مبتنی بر LLM :

@ai_python

👉 Record and understand the LLM response generation process.

👉 Compare many LLM responses in a user-friendly table.

👉 Log traces during LLM development and production.

👉 Use built-in LLM judges to detect hallucinations.

👉 Test the LLM pipeline with different prompts.

👉 Use its pre-configured evaluation pipeline.

https://github.com/comet-ml/opik

معرفی از :

https://substack.com/@avichawla