GitHub Copilot premium requests and AI credits, explained

GitHub Copilot credit capacity pool shrank on September 1, 2026 (by 37% on Business and 44% on Enterprise) while your seat price stayed the same. The first invoice on the new pool lands in early October.

Pool sizes are GitHub’s published figures. The standard pools are 1,900 and 3,900 credits, at 1 AI credit = $0.01 USD². The promotional pools were 3,000 and 7,000 credits from June 1 to September 1, 2026, from the $30 and $70 monthly AI Credit allotments GitHub announced for Business and Enterprise¹.

Seat prices did not change. There is no flat overage rate. Once the included pool is spent, additional usage is billed at each model’s published per-token rate, denominated in AI credits at 1 credit = $0.01⁴. Published rates run $0.10–$20.00 per million input tokens and $0.50–$75.00 per million output tokens³, so what an overage costs depends on which models your org uses, not on a per-unit price you can look up.

Premium requests and AI credits

One billing story, before and after June 1, 2026:

• A premium request was the unit to call to a premium model or agent, chat with a premium model, an agent task, a multi-step edit and weighted by a per-model multiplier⁵. GitHub¹ retired the term: “premium request units (PRUs) will be replaced by GitHub AI Credits.”
• An AI credit is the unit consumption metered on actual tokens which are input, output and cached at each model’s published API rates. A credit corresponds to a cent of that usage, which is how GitHub’s dollar allotments ($19 and $39 per user per month for Business and Enterprise, raised to $30 and $70 during the promotion) become the 1,900 / 3,900 and 3,000 / 7,000 credit pools in the table above.

Under either unit, ordinary code completions and Next Edit suggestions stay included on every plan and consume nothing. Credits are pooled at the billing entity, not metered per seat.

How the credit system is evolving

Under per-seat metering, one developer’s heavy month was one developer’s problem.

June 2026 GitHub replaced premium requests with AI credits and moved organizations and enterprises to usage-based billing (announced April 27, 2026): one pooled monthly credit allowance, metered overage, and a three-month promotional period with much larger pools. Monthly plans migrated automatically and annual plans keep premium-request pricing until renewal. A heavy month became everyone’s problem, and the pool became the thing finance has to forecast.

September 2026 The promotional period ended and pools fell to the standard amounts in the table above. Three important details:

1. Additional usage is enabled by default. Preventing spend beyond the pool requires an administrator to disable the AI credits paid usage policy in AI Controls. Doing nothing is a decision to allow overage.
2. There is no automatic fallback to cheaper models. At the limit, work stops and it does not degrade gracefully.
3. Unused credits are forfeited. No carryover. The pool resets at 12:00 AM UTC on the first of each month.

Why October’s invoice will look different

A forecast built on June through August usage was built against the promotional pool. The same consumption now draws against a pool 37–44% smaller, and everything above the line is billed. Model choice multiplies the effect. Weighting each model’s published Enterprise rate 80% input to 20% output³, the gap between the cheapest and priciest model inside one vendor’s lineup runs from about 2.7× at Google (Gemini 3.8 Flash to Gemini 4 Argon) to 100× at OpenAI (GPT-6 Luna to GPT-6 Astra). The default model can have a large budget impact.

Modeled numbers, not invoices.⁹ An Enterprise seat’s monthly pool falls by 3,100 credits. At 1 credit = $0.01, that is $31 per seat per month that now bills instead of drawing on the pool: $6,200 at 200 seats and $155,000 at 5,000. Both figures assume the organization drew its full promotional pool and keeps consuming at that level. An organization that drew less is exposed proportionally less.

What actually drains a pool

Consumption follows a power law: roughly 20% of users, the ones who adopted agent workflows, draw 70–80% of the credits⁸, and in a shared pool they set the date everyone else runs out.

“Exhausted by day 11 despite lighter usage than previous month.” GitHub Community, after the transition⁷

Five cost saving tips

1. Confirm whether overage is enabled. It is on by default. Turn this off to avoid a surprise budget overage.
2. Re-forecast against the standard pool, not the promotional one. Take August’s actual consumption and divide by the new pool size.
3. Check your default model. On Enterprise rates, the priciest model in a vendor’s lineup can cost 2.7–100× the cheapest per token. GPT-5.6 Sol, for example, runs 17.6× GPT-5.4 nano. Choosing a cheaper default is one of the few levers that works without changing anyone’s behavior.
4. Look at the distribution, not the average. The usage dashboard in Copilot settings and the usage metrics API⁶ will show what the top decile consumes. The mean will tell you almost nothing.
5. Decide what happens when someone hits the wall before the month resets. There is no graceful degradation. Either you have a process, or your engineers have a workaround.

The part nobody has solved

Every native control, the AI credits paid usage policy, additional usage budgets, per-user budgets is a ceiling. Ceilings stop spending and none of them moves unused allocation to the developers who ran out. Both results are true in the same pool, in the same month: forfeiture on one side, blocked engineers on the other. We wrote up what each control does and where it stops for anyone choosing a cap height this month.

We are building Spillway Ops for that missing route: allocation that would otherwise be forfeited flows to the developers who ran out, so spend falls without anyone losing a tool. It reads usage metadata and sets per-user limits through vendor admin APIs with no repository access, no prompt contents, no code. If the October invoice is about to be a conversation at your company, we would genuinely like to hear how it goes.

Before you act on any of the tips above, look at your own pool’s figures. Which lever is worth pulling depends on who in your organization is drawing the credits and how fast.