Datasette commit frequency chart shows coding agents and models drive development surge

Simon Willison, out of curiosity, sought visual evidence of how coding agents and Opus 4.5-level models affected his own output. He found a key metric: the GitHub commit frequency chart for his open-source project Datasette. The chart shows a massive activity spike near the end, coinciding with the release of models like Opus 4.8, GPT-5.5, Fable 5, and GPT-5.6 Sol. This evidence suggests that coding agents and large models significantly boost developers' code output frequency. For developers using the Claude API or similar models, this chart offers a way to quantify personal productivity gains. Teams running local open-source models can also use it to evaluate the real-world benefits of integrating advanced coding agents. It provides concrete data, not just theory, proving the impact of AI-assisted coding tools on actual projects.