1  """LLM-specific features for evaluating language model outputs.
 2  
 3  This module provides tools for working with LLM evaluations, including:
 4  - Prompt templates and rendering
 5  - RAG (Retrieval-Augmented Generation) utilities
 6  - LLM optimization tools
 7  - Data generation for LLM testing
 8  
 9  Most LLM evaluation functionality is available through `descriptors` (e.g.,
10  `LLMEval`, `LLMJudge`) and standard `Report` workflows. This module contains
11  lower-level utilities for advanced use cases.
12  
13  For LLM evaluation workflows, see:
14  - `evidently.descriptors` for LLM-based descriptors
15  - `evidently.presets.TextEvals` for summarizing text evaluations
16  """