# PureAI

## Docs

- [Welcome to PureAI](https://docs.puredocs.org/introduction.md): Complete platform for AI application development
- [Exceptions](https://docs.puredocs.org/lunar/api/exceptions.md): Error handling and exception types in Lunar SDK
- [API Reference](https://docs.puredocs.org/lunar/api/reference.md): Complete API reference for the Lunar SDK
- [Built-in Scorers](https://docs.puredocs.org/lunar/evals/built-in-scorers.md): Pre-instantiated scorers for common evaluation tasks
- [Custom Scorers](https://docs.puredocs.org/lunar/evals/custom-scorers.md): Create your own scoring functions
- [Factory Scorers](https://docs.puredocs.org/lunar/evals/factory-scorers.md): Parameterized scorers that you configure for your needs
- [Evaluations Introduction](https://docs.puredocs.org/lunar/evals/introduction.md): Test and measure the quality of your LLM outputs
- [LLM Judge](https://docs.puredocs.org/lunar/evals/llm-judge.md): Use AI to evaluate LLM outputs
- [Running Evaluations](https://docs.puredocs.org/lunar/evals/running-evals.md): Learn how to run evaluations with the Lunar SDK
- [Async Usage](https://docs.puredocs.org/lunar/guides/async-usage.md): Use AsyncLunar for high-performance asynchronous applications
- [Authentication](https://docs.puredocs.org/lunar/guides/authentication.md): Configure API Key authentication for Lunar SDK
- [Chat Completions](https://docs.puredocs.org/lunar/guides/chat-completions.md): Create chat completions with the Lunar SDK
- [Cost Tracking](https://docs.puredocs.org/lunar/guides/cost-tracking.md): Monitor per-request costs and token usage
- [Fallbacks](https://docs.puredocs.org/lunar/guides/fallbacks.md): Configure automatic fallback models for reliability
- [Models & Providers](https://docs.puredocs.org/lunar/guides/models-providers.md): List available models and providers, and force specific providers
- [Streaming](https://docs.puredocs.org/lunar/guides/streaming.md): Stream responses token by token for real-time output
- [Supported Models](https://docs.puredocs.org/lunar/guides/supported-models.md): Complete list of all supported models with IDs and usage examples
- [Text Completions](https://docs.puredocs.org/lunar/guides/text-completions.md): Create text completions with the Lunar SDK
- [Installation](https://docs.puredocs.org/lunar/installation.md): Complete installation and configuration guide for Lunar SDK
- [Lunar SDK Overview](https://docs.puredocs.org/lunar/overview.md): Python SDK for PureAI LLM Inference API with intelligent fallbacks and built-in evaluations
- [Quickstart](https://docs.puredocs.org/lunar/quickstart.md): Get started with Lunar SDK in 5 minutes
- [Billing](https://docs.puredocs.org/pricing/billing.md): Credit management, payments, and billing details
- [Instance Tiers](https://docs.puredocs.org/pricing/instance-tiers.md): GPU instance types, specifications, and pricing
- [Pricing Overview](https://docs.puredocs.org/pricing/overview.md): Understanding PureAI pricing and billing
- [ChunkCount](https://docs.puredocs.org/purecpp/chunks/chunkcount.md): Split text based on a count pattern.
- [ChunkDefault](https://docs.puredocs.org/purecpp/chunks/chunkdefault.md): The **ChunkDefault** module splits large pieces of text into manageable chunks, using overlap to maintain context between segments. This is particularly useful in **Retrieval-Augmented Generation (RAG)** pipelines and other text processing tasks where continuity matters.
- [ChunkQuery](https://docs.puredocs.org/purecpp/chunks/chunkquery.md): Split text into chunks and filter them based on similarity to a query.
- [ChunkSimilarity](https://docs.puredocs.org/purecpp/chunks/chunksimilarity.md): Split text into chunks and sort them based on similarity.
- [Introduction - Chunks](https://docs.puredocs.org/purecpp/chunks/introduction.md): Chunking modules split large pieces of text into smaller, manageable segments. Overlapping helps maintain context between chunks, making them essential for **Retrieval-Augmented Generation (RAG)** pipelines and other text-processing tasks.
- [Content Cleaner](https://docs.puredocs.org/purecpp/cleaner/cleanregex.md): This module allows cleaning document content using regex patterns, removing unwanted characters, extra whitespace, or other artifacts before further processing.
- [Embedding](https://docs.puredocs.org/purecpp/embbeding/embbeding.md): Generate text embeddings using OpenAI's embedding model.
- [DOCX Loader](https://docs.puredocs.org/purecpp/loaders/docxloader.md): This data loader allows loading DOCX files from local storage.
- [Introduction - Data Loader](https://docs.puredocs.org/purecpp/loaders/introduction.md): Data loaders do PureCPP convertem dados brutos em um formato padronizado, garantindo consistência entre diferentes fontes de dados. Cada loader segue uma estrutura unificada, oferecendo um conjunto consistente de métodos e uma experiência de uso contínua.
- [PDF Loader](https://docs.puredocs.org/purecpp/loaders/pdfloader.md): This data loader allows loading PDF files from local storage.
- [TXT Loader](https://docs.puredocs.org/purecpp/loaders/txtloader.md): This data loader allows loading text files from local storage.
- [WEB Loader](https://docs.puredocs.org/purecpp/loaders/webloader.md): This data loader allows loading webpages from the internet.
- [Metadata Extractor](https://docs.puredocs.org/purecpp/metadata/metadata.md): The **MetadataRegexExtractor** module is designed to extract structured metadata from documents by applying regular expression (regex) patterns. It identifies elements such as proper names, dates, numbers, emails, URLs, and custom patterns.
- [PureCPP Setup](https://docs.puredocs.org/setup.md): Installation and environment configuration guide for PureCPP, an independent product from PureAI

## OpenAPI Specs

- [openapi](https://docs.puredocs.org/api-reference/openapi.json)

## Optional

- [Community](https://discord.gg/8eF9v78Ndv)
- [GitHub](https://github.com/pureai-ecosystem)