This survey provides an in-depth review of the latest state-of-the-art Large Language Models (LLMs) applied in graph learning, introducing a novel taxonomy to categorize existing methods based on their framework design. The paper discusses four unique designs: i) GNNs as Prefix, ii) LLMs as Prefix, iii) LLMs-Graphs Integration, and iv) LLMs-Only. Each category is detailed with representative examples, highlighting key methodologies and exploring the strengths and limitations of each framework. The survey emphasizes potential avenues for future research, including overcoming current integration challenges between LLMs and graph learning techniques, and venturing into new application areas. The paper also discusses the integration of LLMs with graph data, including alignment between GNNs and LLMs, fusion training of GNNs and LLMs, and LLMs-based agents for graph tasks. The survey aims to serve as a valuable resource for researchers and practitioners eager to leverage large language models in graph learning, and to inspire continued progress in this dynamic field. The authors maintain the related open-source materials at https://github.com/HKUDS/Awesome-LLM4Graph-Papers.This survey provides an in-depth review of the latest state-of-the-art Large Language Models (LLMs) applied in graph learning, introducing a novel taxonomy to categorize existing methods based on their framework design. The paper discusses four unique designs: i) GNNs as Prefix, ii) LLMs as Prefix, iii) LLMs-Graphs Integration, and iv) LLMs-Only. Each category is detailed with representative examples, highlighting key methodologies and exploring the strengths and limitations of each framework. The survey emphasizes potential avenues for future research, including overcoming current integration challenges between LLMs and graph learning techniques, and venturing into new application areas. The paper also discusses the integration of LLMs with graph data, including alignment between GNNs and LLMs, fusion training of GNNs and LLMs, and LLMs-based agents for graph tasks. The survey aims to serve as a valuable resource for researchers and practitioners eager to leverage large language models in graph learning, and to inspire continued progress in this dynamic field. The authors maintain the related open-source materials at https://github.com/HKUDS/Awesome-LLM4Graph-Papers.