Large Language Models (LLMs) have demonstrated significant progress in supporting various stages of the software development lifecycle, particularly in code generation, analysis, and testing. These models are capable of synthesizing source code from natural language and assisting in the generation of test cases, thereby enhancing development efficiency, code quality, and automation. In the context of security testing, LLM-aided systems can generate fuzz targets from large codebases and perform continuous fuzzing to identify potential vulnerabilities. However, there are still challenges in fully leveraging these solutions, especially in resource-constrained environments. This paper presents an experimental study on the application of LLMs for generating fuzz targets for library functions under limited computational and data availability, highlighting both the strengths and limitations of this approach.