Microsoft's GitHub shifts to metered AI billing amid cost crisis
GitHub Copilot's free ride is over. Microsoft shifts to metered billing as AI inference costs bite.

Why it matters
GitHub Copilot, one of the highest-volume AI inference endpoints in production, is moving from unlimited to metered pricing. This signals that even at massive scale, per-token economics of LLM inference are forcing product model changes—and hints at broader margin pressure across the AI-as-a-feature market.
The key facts
5 to knowGitHub Copilot moving from unlimited/all-you-can-eat to metered billing model
Cost crisis cited as driver for pricing change
Impacts millions of developers using Copilot daily
Published April 28, 2026—recent implementation
Signals inference cost pressure at production scale
Go to the source
The Register AI/MLtheregister.com
Publisher excerpt: The all-you-can-eat AI buffet is coming to an end