“Even though I was very cautious on the first day and tested with a limited number of uses, it still spent 840 credits,” said one user he wrote Testing Claude Sonnet 4.6 via Copilot today. “I haven’t done anything complicated yet,” complained another user informed usage representing 21 percent of the credit share of a monthly Pro Copilot subscription in one day. “I have a feeling I’ll be going somewhere else soon.”
Using all 8,000 of your organization’s monthly AI credits in one day is…probably not sustainable.
In light of the price change, many GitHub Copilot users are predictably and publicly threatening to cancel their subscriptions or look for other AI coding options. But others say they can adapt to the new world of usage-based pricing. Coder Henry Kinnunen writes Using Claude 5.3-Codex via Copilot they only burned 161 credits “in a productive day”, as they were limited to “very careful and deliberate changes with the AI”. At Bluesky, coder Neil Hewitt he wisely noted Continuing a three-day chat session on Copilot probably isn’t very wise now, because it means “sending the entire chat history as context every time… hey, access tokens use credits… it’s not rocket science”.
Although some Copilot users are jumping ship for other services with more generous usage limits, this type of subsidized customer purchase may soon give way to Copilot-style usage-based pricing in the industry. If this happens, LLMs who are more efficient with their tokens may win the economic battle; There is a user on reddit is already discussed How they integrated Deepseek into their GitHub VSCode environment for only “about 7 cents for 15 million tokens”. While you might say “you get what you pay for,” some AI users are now envisioning a world where they have to pay for what they get.
