GitHub Octoverse 2026
Simon Willison’s Weblog Simon Willison
GitHub says AI agents sent activity soaring: merged pull requests hit 220 million a month. Commits and new repos also jumped hard, which is a pretty loud signal from one year to the next.
Based on reporting by Simon Willison’s Weblog, Simon Willison — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
GitHub used its Universe stage to put a number on something a lot of people have been talking around: agentic engineering is showing up in the stats. EVP Jay Parikh unveiled this year’s Octoverse report and singled out the biggest changes in platform activity.
The headline figures are hard to miss. Merged pull requests tripled in less than a year to 220 million a month. Commits more than tripled to 7.38 billion a month, and GitHub says most of that rise landed in the second half of this year. New repository creation also doubled, reaching 30.3 million per month.
And the report is not just a top-line victory lap. It adds industry breakdowns and a separate look at repositories connected to interoperability or reverse engineering. That means GitHub is trying to map where the growth is happening, not just how much of it there is.
The methodology notes are doing some heavy lifting here. For country data, GitHub summed pushes to public repositories over the Octoverse year and adjusted for VPN noise by using LLM classification to pull the ISO 3166-1 alpha-2 code from user profiles. For industry, it used a fine-tuned multilingual embedding model to sort public repositories created from 2022 to 2026 into NAICS sectors using names, descriptions, topics and READMEs. For interoperability, it counted repositories whose public fields mention terms like clean room implementation, drop-in replacement or protocol re-implementation.
My take — AI-written commentary, not fact-checked reporting
This is the kind of report that makes the agent hype look less like a demo and more like a production shift. GitHub is basically admitting that the bots aren’t just chatting anymore; they’re moving the numbers. The bigger question now is whether developers are steering the agents, or just cleaning up after them.
Read more about this at: Simon Willison’s Weblog