AA Coding Agent Index

Artificial Analysis' overall score for coding agents, combining three coding benchmarks with what each run costs and how many tokens it burns. It rates a model paired with a particular agent harness rather than the model alone, so the same model scores differently in different tools. Higher is better.

GPT-6 Astra by OpenAI holds the best AA Coding Agent Index result among tracked models, at 67 (released Sep 3 2026). Higher scores are better on this benchmark.

1. GPT-6 Astra (OpenAI) — 67; 2. Mistral Large 4 (Mistral) — 49.8.

2 tracked models have a published AA Coding Agent Index score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.