Sorry, you have been rate-limited. Please wait
The message
Sorry, you have been rate-limited. Please waitWhat it means
GitHub's Copilot service refused the request because a rate limit was hit. The Error Code line under the message names the limit, and the wait time in the first line is the server's own estimate.
What to do
Read the Error Code. For a per-model limit, switch to another model or to Auto. For a session or weekly limit, wait for the time it gives. Update VS Code, since newer builds word each limit more clearly.
This is the full block from an April 2026 report by a Copilot Pro+ subscriber:
Sorry, you have been rate-limited. Please wait a moment before trying again. Learn More
Server Error: Sorry, you've exhausted this model's rate limit. Please try a different model. Please review our Terms of Service.
Error Code: user_model_rate_limited
The first sentence is written by the Copilot Chat code in VS Code (which now ships inside VS Code's own repository). The two lines under it come from GitHub's servers, and they're what tell you which limit it was. We read how the message is chosen in VS Code 1.139.1, and there's a whole family of them. 140 issues on the VS Code tracker quote this one as of September 26, 2026.
Reading the Error Code
Current builds turn the common codes into their own sentences. The generic "you have been rate-limited" line is what's left over: a code the extension doesn't special-case, or an older VS Code that didn't have the newer wordings yet. The codes you'll run into:
user_model_rate_limited: you've used up the limit for this one model. Other models still work. Newer builds say "You've hit the rate limit for this model."user_global_rate_limited: a limit across Copilot as a whole, shown as "You've hit your session rate limit" in newer builds. On individual plans (the code checks forfree,individualandindividual_pro) it also suggests upgrading.user_weekly_rate_limited: the weekly allowance is gone. Newer builds give the reset date.agent_mode_limit_exceeded: agent mode has its own limit, and the message says to switch to ask mode (plain questions and answers, no tools).model_overloadedorupstream_provider_rate_limit: the model's provider is busy. This one has nothing to do with how much you've used.
The wait in the first line is a retry time the server sends back. When it sends none, the code fills in "a moment", so that phrase tells you nothing about how long the limit lasts. In one April report the line said "Please wait 2 minutes" with user_weekly_rate_limited under it, and the block was still there an hour later (the reporter says their weekly limit hadn't reset on schedule).
Why the message says “consider switching to Auto”
When you've picked a model yourself, the message adds "or consider switching to Auto". Auto hands the choice of model to Copilot. GitHub's troubleshooting page says "most people see rate limiting for select models, due to limited capacity", so for the per-model and busy-provider codes, Auto or another model is the quickest way back. It won't help with a weekly limit, which covers your account.
When it doesn't clear
GitHub's page says the service-level limits "should not affect typical or even deeply engaged Copilot usage", and it points anyone who keeps getting limited to GitHub Support. The VS Code team closes these reports as duplicates of its tracking issue, and the April one above was also told its VS Code was out of date. Updating first gets you the clearer wording. The standalone Copilot CLI (a terminal tool, separate from the editor) has its own version of the weekly message, covered on our “You've reached your weekly rate limit” page.
Other lines the same feature prints
Match yours against these if the one at the top of the page is not quite it. They come from the same code and mean related things.
Sorry, you have been rate-limited. Please wait a moment before trying again.Sorry, you have been rate-limited. Please wait before trying again or consider switching to Auto.Sorry, your request was rate-limited. Please wait a moment before trying again.Server Error: Sorry, you've exhausted this model's rate limit. Please try a different model.Error Code: user_model_rate_limitedError Code: user_weekly_rate_limited