Quick Start

Quick Start - Choose Your Path

GLChat supports two ways to create custom AI pipelines. Choose the one that fits your needs

GLChat Pipeline Options

🔍 Pipeline Comparison Overview

Category	Custom Internal Pipeline	External Pipeline
Integration	Built inside GLChat backend	Connected via webhook
Code Changes	Requires modifying GLChat codebase	No GLChat code changes
Deployment	Requires GLChat redeployment	Independent deployment
Performance	Higher (no network calls)	Network latency involved
Infrastructure Access	Full access to GLChat components	Limited to API capabilities
External Dependency	No	Yes
Setup Complexity	Higher	Lower
Best For	Deep integration & complex workflows	Fast integration & external models

---

🌐 Option A: External Pipeline (Webhook Integration)

Connect your external AI service via webhook

Integrate an external AI service that you host separately. GLChat sends requests to your service via webhook, and your service returns AI responses.

Best for:

• Using existing AI services you've built

• Integrating fine-tuned models hosted externally

• Quick integration without modifying GLChat code

✅ Pros:

• No GLChat code changes needed

• Register in minutes

• Works with any OpenAI-compatible service

• Independent deployment and scaling

• No GLChat redeployment required

❌ Cons:

• External service dependency

• Network latency

• Limited to OpenAI Responses APIs

• Requires managing external service

🎯 Use when:

• You have a custom fine-tuned model

• You want to use a specialized AI service

• You need to integrate external AI providers

• You want to test models quickly

• You don't want to modify GLChat codebase

---

🔧 Option B: Custom Internal Pipeline (Internal Codebase)

Add your pipeline directly into GLChat's codebase

Build custom AI workflows by writing code within GLChat. Your pipeline becomes part of the GLChat backend and uses GL SDK components directly.

Best for:

• Complex custom workflows

• Deep integration with GLChat features

• Using GLChat's internal components (retrievers, synthesizers, etc.)

• Full control over pipeline logic

✅ Pros:

• Full access to GLChat infrastructure

• No external service dependencies

• Better performance (no network calls)

• Complete control over pipeline behavior

❌ Cons:

• Requires coding knowledge

• Must redeploy GLChat for changes

• More complex setup process

🎯 Use when:

• You need custom retrieval logic

• You want to integrate with GLChat's components

• You're building domain-specific workflows

• You need tight integration with GLChat features

---

🚀 Get Started

Choose your path and follow the step-by-step guide:

• 🌐 Option A: External Pipeline → Guide Connect your external AI service via webhook

• 🔧 Option B: Custom Pipeline → Guide Build your pipeline inside GLChat's codebase

---

---