Research report·
AI Coding Tools Benchmark
How AI coding tools compare in terms of key performance indicators
Artificial Intelligence (AI) has now become a core infrastructure component of modern software engineering, reshaping everything from how code is written to how software teams deliver value. In this report, we benchmark the rapidly evolving landscape of AI coding assistants, agents, and AI-native IDEs – otherwise referred to as AI coding tools. By doing this, we provide a clear path to understanding not just which tools are leading, but how they are fundamentally changing how developers work. The goal is to offer buyers clarity on where AI delivers measurable impact, and to highlight for vendors where the strongest opportunities and gaps exist.
This report presents data from our inaugural AI Coding Tools Benchmark. The results were collected in Q4 2025 from a global panel of 837 professional developers who use AI coding tools. We benchmarked 16 of the most prominent AI coding tools, selected for their market impact and technological relevance. They are, in alphabetical order: Aider, Amazon Q Developer, Claude Code, Cline, Cursor, Firebase Studio, Gemini Code Assist, GitHub Copilot, GitLab Duo, JetBrains AI, Mistral Code, OpenAI Codex, Replit, Sourcegraph Amp, Tabnine, and Windsurf.