What to do when Antigravity hits its quota: 9 ways to keep coding in 2026

Disclosure: Enjoy is our product, and it runs Antigravity. We checked every price and limit below against Google’s, Anthropic’s, OpenAI’s and OpenCode’s own pages on November 3, 2026, and we say where another way is the better pick, including the free ones that don’t need Enjoy.

The ways to keep coding, compared

Antigravity’s quota belongs to your Google account and your Google AI plan. Google’s plans page gives accounts without Pro or Ultra “A meaningful quota, refreshed weekly,” Google AI Pro “A high, generous quota, refreshed every five hours until the weekly limit is reached,” and Ultra “The highest, most generous quota, refreshed every five hours,” with “The highest weekly rate limits.” It doesn’t publish the sizes, and it adds: “Usage limits for this service are subject to modification.” What to do depends mostly on your plan and how far off the refresh is.

Way	Cost, checked November 3, 2026	How soon you’re back	Best for
Wait for the refresh	Free	Five hours on Pro and Ultra; up to a week on free accounts or at a weekly limit	A refresh that’s close
Switch to the other group of models	Free	Right away, if that group has quota left	Pro and Ultra, when Claude or Gemini runs out first
AI credits	Bought from Google One, in amounts such as 2,500, 5,000 or 20,000 credits	Right away, with AI Credit Overages on Always	Finishing today’s task on Pro or Ultra
Upgrade to Pro or Ultra	Google AI Pro $19.99 a month; Ultra $99.99 or $199.99	Right away	Running out most weeks
A Gemini API key, in Antigravity CLI	Gemini 3.8 Flash free on the API’s free tier; paid, $0.75 in and $3.75 out per million tokens	Right away, in the CLI	Gemini without your account’s quota
Claude Code, on a Claude plan	Claude Pro $20 a month; Max $100	Right away, in a new conversation	Claude on Anthropic’s own limits
Codex, on a ChatGPT plan	Free; Go $8 a month; Plus $20; Pro from $100	Right away, in a new conversation	A second agent at no cost
OpenCode, with free or local models	Free	Right away, in a new conversation	No bill and no plan to run out
Enjoy	A monthly subscription, free for 7 days, with your phone and other computers (pricing)	Right away, in the same conversation	Handing the conversation to another agent

Prices are each company’s US list prices, before tax. Waiting, switching models, Gemini 3.8 Flash on the API’s free tier, Codex on ChatGPT’s Free plan and OpenCode’s free and local models cost nothing extra.

Wait for the refresh

Best for: a five-hour refresh on Pro or Ultra, or a weekly one that’s a day away.

Antigravity shows what’s left in its model picker, under View Usage, and Google’s plans page says “You can view baseline quota usage across models on the Settings page.” In Antigravity CLI, /usage, or its alias /quota, “refreshes your model configuration and quota status from the backend” and opens a Model Quotas panel, as its docs describe.

How fast you get there depends on the work. Google says its limits “correlate with the amount of work done by the agent,” so “you receive more prompts when your tasks are straightforward and the agent can complete the work quickly, while more complex tasks consume more quota.”

Price: free.

Watch for: on a free account, the quota refreshes weekly, so a limit you hit early in the week can mean days without Antigravity. Antigravity CLI waits out short rate limits but not long ones: version 1.2.13, per its changelog, “Adopts server-requested retry delays for API rate limits while halting early on extended caps.”

Switch to the other group of models

Best for: Google AI Pro and Ultra, when Claude or Gemini runs out before the other.

Antigravity’s model picker, as its models page shows it, counts “Gemini Models” and “Claude and GPT models” separately, each with a “Weekly Limit” and a “Five Hour Limit.” When Claude is out, Gemini 3.8 Flash or Gemini 4 Argon may still have quota, and the other way round. On paid Pro, Claude Sonnet 5.5 and Opus 5.5 are “Available on Google AI Pro for non-trial subscriptions only,” and Ultra has them too.

Price: free.

Watch for: free accounts have little to switch to. On Free and Google AI Plus, Claude Sonnet 4.6, Claude Opus 4.6 and GPT-OSS-120b are now marked “Will be removed on November 18, 2026,” a date Google moved from November 2, which leaves Gemini 3.8 Flash, as our memo on the change explains. And a switch waits for the turn in progress: “If you change the reasoning model while the agent is running, it continues to use the previously selected reasoning model until it completes its steps for that turn.”

Turn on AI credits

Best for: finishing today’s task on Google AI Pro or Ultra without changing plans.

“Users on Google AI Pro or Ultra plans can use purchased AI credits (or any one-time promotional credits) for additional overage usage above the baseline quota,” says Google’s plans page. The AI Credit Overages setting decides what happens when a model’s quota runs out: Never means “wait until the baseline quota refreshes before using this model further,” and Always means “Always use AI credits when the baseline quota is exhausted (automatically switches back to using the baseline quota once it refreshes).” In Antigravity CLI, the setting is Use AI Credits, or useG1Credits, and Google’s settings docs say it’s “Off by default.”

To buy credits, Google’s help center says to open the AI credits page in Google One, select Get AI credits and choose an amount, “e.g., 2,500, 5,000, or 20,000 AI credits.” You can set up auto-reload after your first purchase.

Price: “AI credits are consumed at standard Gemini Enterprise consumption pricing.” Google’s help pages don’t list a price per pack. When Google opened AI credits to Antigravity in March, The Register reported $25 for 2,500, and that “Exactly what a credit is worth when used with Antigravity is not described.”

Watch for: “AI credits are not available for purchase for Google AI Plus plans” or in Japan, per the same help page, and members “subscribed through a third party, Pixel Pass, or Apple billing” can’t buy them, per Google’s credits page. In a family group, credits are pooled, and “only the plan manager can buy additional AI credits.” “AI credits may expire after a certain period, as set out when you acquire them.”

Upgrade to Google AI Pro or Ultra

Best for: running out most weeks, not once in a while.

Google’s plans give Google AI Pro, at $19.99 a month, “Entry rate limits to agent model in Google Antigravity.” Ultra starts at $99.99 a month, with “5x higher usage limits vs. AI Pro” and “Higher rate limits” in Antigravity, and $199.99 buys “20x higher usage limits vs. AI Pro” and “Highest rate limits.” Pro’s quota refreshes every five hours “until the weekly limit is reached”; Ultra’s every five hours, under weekly limits of its own.

Price: Pro $19.99 a month; Ultra $99.99 or $199.99.

Watch for: a trial isn’t the same. The Claude 5.5 models are for “non-trial subscriptions only,” and Gemini 4 Argon is “Available with paid Google AI Pro plans or on trial with eligible premium devices”; to switch from a trial to paid Pro, Google’s FAQ says to “Submit a request to cancel your trial.” Google publishes no size for any plan’s quota. And Google AI Plus, at $4.99, doesn’t change Antigravity’s: the models page puts “Free & Google AI Plus” in one column, and the plans page gives everyone “not on Google AI Pro or Ultra plans” the same weekly quota.

A Gemini API key, in Antigravity CLI

Best for: keeping Antigravity’s agent, with Gemini billed by the token instead of your account’s quota.

Google’s authentication docs say you can “Run Antigravity CLI with your own Gemini API key instead of a signed-in Google account. Model requests go directly to the Gemini API, and the CLI never establishes an account session.” Create a key in Google AI Studio, set "modelProvider": "gemini" in ~/.gemini/antigravity-cli/settings.json, export the key as GEMINI_API_KEY, and run agy. Both steps matter: “Only setting a GEMINI_API_KEY environment variable on its own has no effect.”

Price: on Google’s Gemini API pricing, Gemini 3.8 Flash is “Free of charge” on the free tier. On the paid tier it’s $0.75 per million input tokens and $3.75 per million output through December 31, 2026, then $1.50 and $7.50. Gemini 4 Argon “is not available on the Free tier,” and costs $2 and $10 through November 30, 2026, then $4 and $20.

Watch for: on the free tier, your content is “Used to improve our products,” per the same page, and the paid tier’s isn’t. Free or paid, the API has limits of its own: “Rate limits are applied per project, not per API key,” and daily ones “reset at midnight Pacific time,” per its rate limits page. Google documents the key for Antigravity CLI, saying “This suits headless and CI runs,” not for the Antigravity App, and it brings Gemini only, not Claude.

Claude Code, on a Claude plan

Best for: Claude, on Anthropic’s own plan and limits, when you’d rather not pay Google for it.

Claude Code is Anthropic’s coding agent. Claude’s pricing says Claude Code “is included in all paid plans,” and its usage is Anthropic’s, separate from Google’s quota.

Price: Pro at $20 a month, or $17 a month billed yearly ($200 up front), and Max at $100. The free plan leaves Claude Code out.

Watch for: Anthropic’s limits replace Google’s: “Pro and Max plans have a five-hour session limit and a weekly limit,” and they’re “shared across Claude and Claude Code,” says its help center. Claude Code doesn’t see your Antigravity conversation, so start it with where things stand. Our Claude Code usage limit guide covers what to do when that runs out too.

Codex, on a ChatGPT plan

Best for: a second agent at no cost, on a plan you may already have.

“ChatGPT Work and Codex are included in your ChatGPT Free, Go, Plus, Pro, Business, Edu, or Enterprise plan,” says Codex’s pricing. Free and Go get “GPT-6 Luna at Standard speed in the desktop app, subject to rollout,” and Plus adds “Codex on the web, in the CLI, in the IDE extension, and on iOS” with GPT-6.1 Sol.

Price: Free is $0, Go $8 a month and Plus $20, and Pro comes in “Plans at $100, $200, or $500 USD per month.”

Watch for: OpenAI publishes estimates, not limits. Its message counts per five hours “are not fixed message limits,” “Weekly limits may also apply,” and “ChatGPT Work and Codex share usage.” Like Claude Code, it starts a new conversation. Our Codex usage limit guide covers its limits.

OpenCode, with free or local models

Best for: routine work with no bill and no plan to run out.

OpenCode is an open-source coding agent, as a terminal interface, desktop app and IDE extension. Its homepage says “OpenCode is 100% free to use. It also comes with a set of free models,” and its docs add “You can configure opencode to use local models through Ollama.” Our local model guide covers the setup and the memory it takes.

Price: free, with its free models or a model on your own computer.

Watch for: its free models are free “for a limited time,” and for several of them, “The team is using this time to collect feedback and improve the model.” Don’t sign it in with your Antigravity login: Google’s FAQ names OpenCode, Claude Code and OpenClaw, and says “Using third-party software, tools, or services to access Antigravity is a violation of our Terms of Service,” which “can result in suspension or termination of your account.”

Enjoy

Best for: keeping Antigravity, with Claude Code and Codex ready to take over the same conversation when its quota runs out.

Enjoy runs Antigravity on your Google account, along with Claude Code, Codex, Grok Build, Muse Code, and local models through OpenCode, on Mac, Windows and Linux, each on its own lab’s plan. When Antigravity runs out:

• You see it coming. Enjoy shows “the limits Google shows for each group of models, such as Gemini · Weekly,” and when each resets.
• The conversation waits and carries on. Enjoy waits until the limit resets, then tries again, “and the agent continues where it left off,” a weekly limit included.
• Or another agent picks it up. Select Stop waiting, choose Claude Code, Codex or OpenCode, and the new agent “gets the conversation so far, including replies from other agents,” as our guide to switching agents puts it.

Price: Enjoy is one monthly subscription, and the first seven days are free, with a card to start. It includes your phone and any browser and your other computers, and each teammate you invite needs a subscription of their own. See pricing.

Watch for: Enjoy doesn’t add to Google’s quota. Antigravity in Enjoy uses your Google account’s quota and runs “the Gemini models Google offers your account,” so for Claude through Google, use Google’s own app. It uses one Google account in every project, so Enjoy can’t move Antigravity to a second account the way it does for Claude Code, Codex and Grok Build. Waiting needs “Your computer must be on, with Enjoy open, when the usage resets.” Enjoy runs Antigravity through the server Google offers “for Xcode, JetBrains IDEs, Zed, and custom ACP clients,” as its changelog puts it, not by wrapping agy. Enjoy isn’t open source, and it’s a monthly subscription on top of your AI plans once the free week ends. If Antigravity is the only agent you use, the free ways above may be all you need.

When to stay on Antigravity and wait

If you’re on Pro or Ultra and the refresh is a few hours off, waiting is free. If you run out once a month, a bigger plan is money spent for a rare afternoon: switch groups, or pay for that day with AI credits. Upgrade when it happens most weeks. On a free account, Gemini 3.8 Flash on a free API key is the cheapest way to keep Antigravity’s own agent working until the weekly refresh.

Skip the “quota fixes” that sign other tools in with your Google account or rotate accounts for you. Google’s FAQ says “Wrapping agy inside third-party coding agents or applications, or automating account rotation, is not supported under consumer plans,” and that using other software with your Antigravity login can get the account suspended. A plan, credits, an API key or another lab’s agent gets you back to work without putting the account at risk.

How we compared

We read Google’s Antigravity plans, models, pricing, FAQ, authentication, CLI and changelog pages, its Google AI plans and AI credits help pages, the Gemini API’s pricing and rate limits, Anthropic’s and OpenAI’s pricing and help pages, and OpenCode’s homepage and docs on November 3, 2026, and every one is linked below. We quote each company’s own words for its limits, leave out what it hasn’t documented, and name the publication when a fact comes from a report. Plans and limits change often, so we recheck this guide against its sources at least once a month, and the date at the foot of the page says when we last did.