> ## Documentation Index
> Fetch the complete documentation index at: https://docs.athina.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Preset Evals

Preset evaluators are a set of common turnkey evaluators that you can use to evaluate your LLM applications.

<Tip>
  You can also create custom evaluators. See [here](/evals/custom-evals) for
  more information.
</Tip>

#### RAG Evals

These evals are useful for evaluating LLM applications with Retrieval Augmented Generation (RAG):

* [Context Contains Enough Information](/api-reference/evals/preset-evals/rag/context-sufficiency)
* [Does Response Answer Query](/api-reference/evals/preset-evals/rag/answer-completeness)
* [Response Faithfulness](/api-reference/evals/preset-evals/rag/response-faithfulness)
* [Groundedness](/api-reference/evals/preset-evals/rag/groundedness)

#### RAGAS Evals

[RAGAS](https://docs.ragas.io/en/stable/) is a popular library with state-of-the-art evaluation metrics for RAG models:

* [Context Precision](/api-reference/evals/preset-evals/rag/ragas#context-precision)
* [Context Relevancy](/api-reference/evals/preset-evals/rag/ragas#context-relevancy)
* [Context Recall](/api-reference/evals/preset-evals/rag/ragas#context-recall)
* [Faithfulness](/api-reference/evals/preset-evals/rag/ragas#faithfulness)
* [Answer Relevancy](/api-reference/evals/preset-evals/rag/ragas#answer-relevancy)
* [Answer Semantic Similarity](/api-reference/evals/preset-evals/rag/ragas#answer-semantic-similarity)
* [Answer Correctness](/api-reference/evals/preset-evals/rag/ragas#answer-correctness)
* [Coherence](/api-reference/evals/preset-evals/rag/ragas#coherence)
* [Conciseness](/api-reference/evals/preset-evals/rag/ragas#conciseness)
* [Maliciousness](/api-reference/evals/preset-evals/rag/ragas#maliciousness)
* [Harmfulness](/api-reference/evals/preset-evals/rag/ragas#harmfulness)

#### Safety Evals

These evals are useful for evaluating LLM applications with safety in mind:

* [PII Detection](/api-reference/evals/preset-evals/safety/pii-detection): Will fail if PII is found in the text
* [Prompt Injection](/api-reference/evals/preset-evals/safety/prompt-injection): Will fail if any known Prompt Injection attack is found in the text
* [OpenAI Content Moderation](/api-reference/evals/preset-evals/safety/open-ai-content-moderation): Will fail if text is potentially harmful
* [Guardrails](/api-reference/evals/preset-evals/safety/guardrails): A popular library for custom validators for LLM applications:
  * [Safe for work](/api-reference/evals/preset-evals/safety/guardrails#sfw): Checks if text has inappropriate/NSFW content
  * [Not gibberish](/api-reference/evals/preset-evals/safety/guardrails#not-gibberish): Checks if response contains gibberish
  * [Contains no sensitive topics](/api-reference/evals/preset-evals/safety/guardrails#contains-no-sensitive-topics): Checks for sensitive topics

#### Summarization Evals

These evals are useful for evaluating LLM-powered summarization performance:

* [Summarization Accuracy](/api-reference/evals/preset-evals/summarization-qa)

#### JSON Evals

These evals are useful for validating JSON outputs:

* [JSON Schema Validation](/api-reference/evals/preset-evals/json-evals#json-schema)
* [JSON Field Validation](/api-reference/evals/preset-evals/json-evals#json-validation)

#### Function Evals

Unlike the previous evaluators which used an LLM for grading, function evals use simple functions to check if:

* Text matches a given [regular expression](/api-reference/evals/preset-evals/function-evals#regex)
* Text [contains a link](/api-reference/evals/preset-evals/function-evals#containslink)
* Text [contains keywords](/api-reference/evals/preset-evals/function-evals#contains-any)
* Text [contains no invalid links](/api-reference/evals/preset-evals/function-evals#noinvalidlinks)
* Text is [missing keywords](/api-reference/evals/preset-evals/function-evals#containsall)

Head over to the [function evaluators](/api-reference/evals/preset-evals/function-evals) page for further details.

#### Evals with Ground Truth

These evaluators compare the response against reference data:

* [Answer Similarity](/api-reference/evals/preset-evals/grounded-evals#answer_similarity)
* [Context Similarity](/api-reference/evals/preset-evals/grounded-evals#context_similarity)

Head over to the [grounded evaluators](/api-reference/evals/preset-evals/grounded-evals) page for further details.