How Tencent’s Large Language Models Transform Business with RAG, GraphRAG, and Agents

Introduction

In this article we explore Tencent's large language model applications across various business scenarios, focusing on how cutting‑edge techniques improve model intelligence and user experience.

Key Topics

Tencent large language model application scenarios

RAG technology principles and practice

GraphRAG in role‑playing scenarios

Agent technology principles and applications

Q&A

Application Scenarios

Tencent's models are used in the WeChat ecosystem, social content, video news, office documents, games, and more, enabling intelligent and efficient solutions.

Content generation: ad copy, comment assistance, etc.

Content understanding: text moderation, fraud detection.

Intelligent customer service: knowledge Q&A, user guidance.

Development Copilot: automated code review, test case generation.

Role‑playing: intelligent NPC interaction in games.

Core Technologies

Tencent employs three main techniques:

(1) Supervised Fine‑Tuning (SFT)

Fine‑tunes a base model with domain‑specific data to embed business knowledge, enabling targeted task responses.

(2) Retrieval‑Augmented Generation (RAG)

Combines external knowledge bases and retrieval with generation, improving explainability and reducing hallucinations; used in intelligent customer service and document assistants.

(3) Agents

Integrates external tools so the model can perform multi‑step reasoning, planning, and execution for complex tasks.

These technologies together drive the practical deployment of AIGC in Tencent’s products, delivering more accurate and efficient AI‑driven solutions.