AI costs how much? GitHub Copilot users react to new usage-based pricing system.

“Even though I was very cautious on the first day, trying it out with a limited number of uses, it still consumed 840 credits,” one user wrote today about testing Cloud Sonnet 4.6 via Copilot. “I haven’t even done anything complicated yet,” complained another user after reporting he used 21 percent of his monthly Pro Copilot subscription’s credit allotment in a single day. “I have a feeling I’ll be moving somewhere soon.”

Amidst the price change, many GitHub Copilot users are predictably and publicly threatening to cancel their subscription or look for other AI coding alternatives. But others say they have been able to adjust to the new world of usage-based pricing. Coder Henri Kinnunen writes that he burned only 161 credits in a “productive day” of using the GPT 5.3-codecs through Copilot, thanks to limiting himself to “very focused and deliberate changes with the AI”. At BlueSky, coder Neil Hewitt wisely notes that maintaining a three-day chat session on Copilot is probably not that wise anymore, as it means “sending the entire chat history as a reference every time… Hey, input tokens use credits… It’s not rocket science.”

While some Copilot users are jumping ship to other services with more generous usage limits, such subsidized customer acquisition may soon pave the way for Copilot-style usage-based pricing across the industry. If this happens, those LLMs who are more efficient with their tokens may win the economic battle; On Reddit, one user is already discussing how they have integrated Deepseek into their GitHub VSCode environment at a cost of only “about 7 cents for 15 million tokens.” While you might say “you get what you pay for,” some AI users are now considering a world where they must also pay for what they get.