This study evaluates leading Large Language Models (LLMs) for intelligent code completion capabilities using a novel evaluation framework.

• Compares performance of Gemini 1.5 (Flash & Pro), GPT-4o, GPT-4o-mini, and GPT-4 Turbo
• Uses the Syntax-Aware Fill-in-the-Middle (SAFIM) data framework for evaluation
• Focuses on context-aware code completion in modern development environments
• Provides actionable insights for selecting appropriate LLMs for software engineering tasks

For engineering teams, this research offers valuable guidance on which AI models can most effectively enhance developer productivity and code quality in real-world scenarios.

Comparative Analysis of Large Language Models for Context-Aware Code Completion using SAFIM Framework