From 1 June, GitHub Copilot bills by what you use. Every plan now draws down a monthly allowance of GitHub AI Credits, consumed by token usage — input, output and cached — at published per-model rates. The change, set out on the GitHub blog, replaces the older system of premium request units. Chat, agent and code-review usage now runs on a meter; code completions remain included and do not consume credits.
Disclosure: RECATOOLS is written with the assistance of Claude, made by Anthropic, which is also named in this article. Readers should weigh that interest.
What actually changes
Headline prices hold. Copilot Pro stays at US$10 a month with US$10 of AI Credits, and Pro+ at US$39 with US$39; Business is US$19 a user with US$19 of credits; Enterprise is US$39 with US$39. Instead of the old premium request units, usage is now measured in credits drawn down per token at each model's API rate. Heavier models cost more per call. A developer who lived comfortably inside a flat plan can now run the meter down faster than expected.
Why developers pushed back
The official announcement thread drew more than 400 comments and close to 900 downvotes, per the GitHub community discussion. The main worry was predictability. A flat fee is easy to budget, whereas a token meter is hard to forecast, especially for engineers who keep an AI agent or chat running through the day. There is also a quieter change: from 1 June, code-review workflows consume GitHub Actions minutes too, so the cost of automation creeps in from more than one direction.
What it signals
The change shows the AI-tools business model catching up with the cost of running the models. Flat-rate AI was always a subsidy, and with usage growing while inference stays expensive, vendors are passing that cost back to users. Because GitHub is large enough to set the tone for the market, Enterprise DNA expects more coding tools to follow Copilot onto a meter.
What to do before the meter bites
Check which model your team defaults to, since the gap between a light and a heavy model is where credits disappear. Watch the first month's consumption before assuming the included allowance is enough, and set spending limits where the plan allows. Nothing about how the tools work has changed, only the way the bill is calculated.
Three months on, the meter spread and the rates fell
In May the expectation was that more coding tools would follow Copilot onto a meter, since flat-rate AI was a subsidy that inference costs would eventually end. The bills did not behave the way that reasoning implied.
Metering did spread, and it moved up a layer. Stripe agreed in August to buy OpenRouter, a gateway that routes enterprise work across roughly 400 models and meters the token spend across all of them. Billing by consumption is becoming infrastructure rather than a pricing choice each vendor makes on its own.
Prices did not rise. OpenAI cut its cheapest model by 80 per cent on 1 August, taking Luna from one dollar to twenty cents per million input tokens three weeks after launch, with Terra down 20 per cent and the flagship unchanged. Anthropic had announced a Sonnet 5 rise for 1 September and withdrew it on 10 August, making the US$2 and US$10 rates permanent. Across the large Western labs the movement over the quarter was downward or flat.
DeepSeek is the exception, and it went further than a price rise. V4 now charges by the clock, with peak and off-peak rates, and the increase lands hardest on the cheapest token category. The peak window sits inside ASEAN working hours.
Why falling rates do not answer the pushback
The 400 comments and 900 downvotes on GitHub's announcement thread were about budgeting rather than price level. Falling rates do not answer that objection.
Cheaper units do not restore a fixed monthly number. A team that could previously budget a flat US$19 per developer now budgets a distribution, and the shape of that distribution depends on which model each engineer's editor defaults to. Cheapest-model price cuts help only if the cheap model is the default, and the gap between the top and the bottom of OpenAI's own ladder widened to 25 to 1 over the same period. That makes the default editor setting more consequential than it was in May.
DeepSeek's time-of-day pricing is the sharper version of the same problem. With a flat per-token rate, a team can at least multiply out the cost in advance. Once the rate changes with the clock, an identical task costs different amounts depending on when an engineer runs it, and for teams working in ASEAN hours the expensive window is the working day.
The default model matters more than it did
The practical suggestion in May was to check which model your team defaults to, and the price moves above have made that more important.
Spending limits now deserve the same attention. GitHub was degraded for about three hours on 17 August, with archive downloads failing at roughly half, and code-review workflows have been drawing on Actions minutes since 1 June. Automation billed by consumption keeps consuming while the pipeline underneath it is failing, and retries against a degraded dependency are billed like any other run. It is worth setting those limits before a team finds out what that combination costs.
Correction, 18 October 2026: An earlier version said Copilot autocomplete now runs on a meter; GitHub says code completions remain included and do not consume AI Credits. A paragraph about a Microsoft model called Project Polaris, which could not be supported, has been removed, and the Copilot Pro plan has been added to the price list.
View author profile → · Editorial policy
