GitHub commits are on pace to hit 14 billion this year, up from 1 billion in 2025

GitHub is growing so fast that Microsoft is reportedly renting computing capacity from Amazon to keep up with it. GitHub's chief operating officer Kyle Daigle said in April that code commits, recorded changes to a codebase, the most basic unit of developer activity, are on pace to reach 14 billion in 2026, up from roughly 1 billion in 2025. That's not incremental growth. That's a 14x jump in a single year, and AI-assisted coding is the reason.

The strain on infrastructure is real enough that Microsoft has started turning to a direct competitor for help. Microsoft is reportedly using AWS to add computing capacity for GitHub, even as the company runs its own Azure cloud and is currently facing an antitrust lawsuit alleging it has used licensing terms to lock customers into Azure over rival clouds. Needing AWS capacity for a flagship product while defending against claims of anti-competitive cloud lock-in is the kind of contradiction that doesn't usually make it into a press release.

The numbers behind the strain go well beyond commits. Nearly 140,000 organizations now use GitHub Copilot, and enterprise subscribers have nearly tripled year-over-year, according to Microsoft's most recent earnings call. GitHub Copilot CLI usage is nearly doubling month-over-month, a pace that outstrips even the broader Copilot growth, since CLI use generally signals more advanced, agentic workflows rather than simple autocomplete.

"GitHub itself is seeing unprecedented growth driven by the proliferation of agentic coding, and we are hard at work to scale and meet this demand." - Microsoft executive, FY2026 Q3 earnings call

What this means: the bottleneck in AI coding right now isn't model quality, it's raw infrastructure capacity to run the agents that quality enables. Models like Claude, GPT, and Gemini have gotten good enough that developers are running dozens of agentic sessions a day instead of occasional autocomplete suggestions, and that volume is straining the platform that hosts the code itself, not just the AI vendors generating it.

The financial picture backs up the infrastructure strain. Microsoft's AI business surpassed a $37 billion annual revenue run rate in its most recent quarter, up 123% year-over-year, with gross margin percentage declining specifically due to continued AI investment and increased GitHub Copilot usage. Microsoft is paying a real margin cost to keep GitHub running at this pace, and choosing to pay it anyway, including, apparently, by paying a competitor for capacity.

This kind of demand surge is exactly the backdrop behind GitHub's recent move to usage-based pricing for Copilot, announced just this week, a shift that lets GitHub capture revenue more directly from the heaviest agentic users, rather than charging every developer the same flat seat price regardless of how many AI sessions they actually run.

Whether AWS capacity is a temporary patch or a sign that GitHub's growth is outrunning even Microsoft's own infrastructure roadmap is the real question heading into Microsoft's next earnings call. If commits really do hit 14 billion this year, the platform handling the world's code just had to scale itself by an order of magnitude in twelve months, and the company running it had to go outside its own cloud to do it.