If you found value in this post, consider following me on X @davidpuplava for more valuable information about Game Dev, OrchardCore, C#/.NET and other topics.
It must've been love, but it's over now...
Maybe. Maybe not.
Today is June 1st, 2026 and Microsoft has transitioned GitHub Copilot to usage based billing. For me, this change is a net negative.
To put into perspective, I attempted a relatively small feature change to add a button that cancelled a long running AI call. After the initial request message plus three (3) iterations, I've consumed some 37.8% of my monthly allotment of AI Credits.
All on the first day of the month.
To clarify, I do not blame Microsoft for the change - they need to do what's necessary to provide GitHub Copilot in a cost effective way for them.
But for me, this usage change has me questioning whether GitHub Copilot is still cost effective for me to keep using. Or at least to keep using in the way I was using it before.
I will be the first to admit that I enjoy great success with vibe coding and profressing code review my work with GitHub Copilot.
I successfully launched a game making platform, Lava Web Games with surprising speed and ease. All with a full-time, demanding job and deep family commitments.
I was only able to do this with the help of GitHub Copilot.
Looking back, I should've guessed that it was too good to be true, at least in the long run.
I couldn't tell what my token usage during the honeymoon phase with GitHub Copilot - so my success with it may very well have been because I was sending way more tokens than I should have.
I'm happy to change how I use GitHub Copilot. Do I need to do more focused requests where I add code block snippets for more granular, surgical changes? I'll do so gladly.
I evey tried to do just that, but I don't have a way to know if I'm more efficient with my AI Credits.
I don't know how I can get back to the way I was cooking with GitHub Copilot before. The primary hold back is I just don't know if my simple request is going to be too costly. I would need more information about my request before I send it.
GitHub Copilot needs to provide me more information about how they are calculating my AI Credits. If I had a way to see each requests Input Tokens / Output Tokens / Cache Tokens, I would be able to make a more informed decission.
Even better, would be an real-time estimator of how many credits my next request would cost me. At least then, I could be more thoughtful about each of my changes.
If you found value in this post, consider following me on X @davidpuplava for more valuable information about Game Dev, OrchardCore, C#/.NET and other topics.