> For the complete documentation index, see [llms.txt](https://gdplabs.gitbook.io/glchat/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://gdplabs.gitbook.io/glchat/developer-documentation/deep-research.md).

# Deep Research

## Overview

**GLChat DeepResearch** is an experience-layer pipeline in the GLChat backend that provides deep research capabilities to users. It handles user-facing behavior, profile management, and result presentation for deep research queries.

### Key Features

1. **Profile-Based Research**: Three user-facing profiles (Essentials, Comprehensive, Internal)
2. **Intelligent Routing**: Automatically determines if deep research is needed
3. **Automatic Fallback**: Comprehensive → Essentials when quota exceeded
4. **Streaming Support**: Real-time progress and results
5. **Multi-tenancy**: Full organization-level isolation
6. **Citation Formatting**: Structured references for UI display

### Architecture Context

1. **GLChat DeepResearch** (this document) — Experience-layer pipeline that defines product behavior
2. **GL SDK DeepResearcher** — Library-level SDK abstraction for backend-agnostic deep research (see [GL SDK DeepResearcher docs](https://gdplabs.gitbook.io/sdk/tutorials/generation/deep-researcher))

This document focuses on **GLChat DeepResearch** and how to use it in the GLChat backend.