Here's How Gemini 4 Argon Usage Is Counted in the Gemini App, Antigravity and the API in October 2026

Where Argon is available

Gemini 4 Argon isn't part of the free offer in any of the three places. Google's AI plans page lists access to Argon in the Gemini app for Google AI Pro and Ultra, with "expanded access" for Pro and "higher access" for Ultra. Antigravity's model docs mark Argon as unavailable on the free tier and Google AI Plus, available on Google AI Pro for non-trial subscriptions only, and available on Ultra and Enterprise. The same page says Argon isn't available in the Antigravity IDE, only in the Antigravity App, the Antigravity CLI and the IDE extensions. On the Gemini API, the pricing page says Argon is not available on the free tier.

How the Gemini App counts usage

The Gemini Apps limits page says the app has compute-based usage limits that factor in the complexity of the prompt, the models and features used, and the length of the chat. The limit refreshes every five hours until the weekly limit is reached. Google sets plan limits as multiples, not counts. Google AI Plus gets 2x the standard limits, Google AI Pro gets 4x, and Google AI Ultra gets 5x or 20x the Pro limits depending on the subscription. Google doesn't publish how many Argon prompts any of those levels covers, and the page doesn't give a separate Argon figure.

The same page ranks what costs the most. A short chat prompt is very low, long ongoing chats and large file analysis are medium, and Deep Research and media generation use more, with video generally the most. Google suggests starting a new chat for a new topic so Gemini doesn't have to re-read a long history. The page also says limits may change without notice, including because of capacity constraints. To check usage, open Settings and then Usage Limits at the bottom left of gemini.google.com. The app also warns when usage gets close to the limit and says when it will refresh.

How Antigravity counts usage

Antigravity's plans page says Google AI Ultra gets the highest quota, refreshed every five hours, and the highest weekly rate limits. Google AI Pro gets a high quota refreshed every five hours until the weekly limit is reached. Users without Pro or Ultra get a quota that refreshes weekly. Google says the rate limits correlate with the amount of work the agent does, so a straightforward task uses less than a complex one, and that the limits are mainly set by available capacity and may be modified.

Argon has its own arrangement on Pro. When Antigravity brought Argon to paid Google AI Pro subscribers on October 16, Beijing time, its official X account said Pro gets "a separate weekly-only Argon quota" on top of the existing Gemini bucket for Gemini 3.8 Flash. Google doesn't publish the size of that quota or of the Ultra quota. In the Antigravity CLI, the usage command, also available as quota, opens a panel with the remaining usage for each model, and the app shows baseline usage across models on its Settings page.

How the Gemini API counts usage

On the API, Argon has no free tier, so every request is billed per token, and rate limits follow the project's usage tier. The rate limits page says limits are set per project rather than per API key, are measured in requests per minute, input tokens per minute and requests per day, and that daily request quotas reset at midnight Pacific time. Google doesn't list Argon's per-minute or per-day numbers on that page. It says the active limits for a project can be viewed in Google AI Studio.

The API also enforces spend-based limits over a rolling 10-minute window: $10 on Tier 1, $50 on Tier 2 and $200 on Tier 3. Linking a billing account gives Tier 1, Tier 2 requires $100 paid and three days since the first successful payment, and Tier 3 requires $1,000 and 30 days. Google says hitting a spend-based limit returns a 429 RESOURCE_EXHAUSTED error. On price, the Standard rate for Argon is $2 per million input tokens and $10 per million output tokens through November 30, 2026, and $4 and $20 from December 1.

Context window and output limit

Our Argon launch coverage, based on Google's model docs, gives Argon a unified context window of 1,048,576 tokens for input and output together. The default synchronous maximum output is 65,536 tokens, and with long decode the output can go up to the full million. In the Gemini App, the context window depends on the plan rather than the model. The limits page lists 32k tokens without a plan, 128k on Google AI Plus, and 1 million on Google AI Pro and Ultra.

What to do when you hit a limit

In the Gemini App, Google says subscribers who reach the limit can continue the conversation with Flash-Lite. Otherwise, the options are to upgrade to a plan with higher limits or wait for the five-hour or weekly limit to refresh. In Antigravity, Google AI Pro and Ultra subscribers can keep working past the baseline quota with purchased AI credits, charged at standard Gemini Enterprise consumption pricing. That depends on the AI Credit Overages setting. With Never, Antigravity waits for the quota to refresh, and with Always, it uses credits and switches back to the baseline quota after the refresh. Antigravity doesn't support bring-your-own-key for extra rate limits. On the API, Google's advice for a spend-based 429 is to wait and retry, reduce expensive requests such as long contexts or outputs, and request a rate limit increase if the limit keeps coming up in normal use. Higher tiers come automatically as spending grows. For Gemini CLI, our guides to Gemini CLI usage limits and to its 429 RESOURCE_EXHAUSTED error cover the details.