POLICY 2026-06-18 2026-06-18
免费层和 Google One 用户的 Gemini CLI 被 Antigravity CLI 取代。付费的 Google AI Pro / Ultra 订阅继续按各自的配额走。
/google — GEMINI CLI
LAST RESET
2026-06-18
POLICY Rule change 2026-06-18
A change to the limit rules themselves, affecting your allowance going forward.
2
Total records
0
Banked resets
48.0d
Avg. interval
48.0d
Longest wait
$ From May 2026 Gemini switched to compute-based 5-hour windows with a weekly cap layered on top.
// ACTIVITY · 52 WEEKS UTC
regular banked policy
MON WED FRI
09101112010203040506070809
// LOG · 2 ENTRIES
newest first POLICY 2026-05-01 2026-05-01
Gemini 不再按提示词条数计量，改成按算力计：每 5 小时刷新一个窗口，外面再套一层每周上限。注意在 agent 模式或 Gemini CLI 里，一次提示可能在内部触发几十个模型请求。
// PLANS
| Plan | Price | 5h gate |
|---|---|---|
| Free（个人 Google 账号） $0 Active 每分钟 60 次、每天 1,000 次模型请求 | ||
| Google AI Pro 约 $19.99/月 Active 每人每天 1,500 次模型请求 | ||
| Google AI Ultra 约 $99.99/月起 Active 每人每天 2,000 次模型请求，另含更多 AI credits | ||
The quota counts model requests — in agent mode a single prompt can trigger dozens internally. Prices are from public reporting.
// WHERE TO CHECK
Gemini CLI
会话里输入 /stats model 看 token 消耗。注意它只统计当前会话，不是账号总量
Google Cloud
Gemini Code Assist 的配额在 Cloud 控制台的 Quotas 页面
订阅页
Google One / Google AI 订阅页能看到当前档位和 AI credits
// FAQ
01 When does the Gemini CLI quota reset?
Before May 2026 it counted prompts per day; since then it uses compute-based 5-hour windows with a weekly cap layered on top. Google has not spelled out how the weekly cap resets — go by what your account shows.
02 Why does one question use several requests?
In agent mode or via the Gemini CLI, a single prompt can trigger dozens of model requests internally. The quota counts model requests, not how many times you hit enter.
03 How does Gemini CLI relate to Antigravity CLI?
From 18 June 2026, the Gemini CLI was replaced by the Antigravity CLI for free-tier and Google One users. Paid Google AI Pro / Ultra subscriptions continue on their own quotas.
04 How much does the free tier get?
Signing in to Gemini Code Assist for individuals with a personal Google account officially gives 60 model requests per minute and 1,000 per day. Google AI Pro is 1,500 per user per day, Ultra is 2,000.
05 How do I check my usage?
Run /stats model inside the Gemini CLI to see token usage — but note it covers only the current session, not your account total. Account-level quota lives in the Google Cloud console under Quotas.