Back to Blog

AI

Google AI Ultra Lite — where does the new tier fit?

Google is preparing an AI Ultra Lite plan to slot between $20 Pro and $250 Ultra. The expected price band is $50-80 per month. Here is who it is for, how it compares to a stacked ChatGPT Plus + Claude Pro subscription, and whether it is worth replacing what you already have.

S
Sarma
8 May 20266 min read
ShareLinkedInX

Google's consumer AI pricing has two extremes — $20 Pro and $250 Ultra. The gap is awkward. Pro caps out fast for anyone doing real work; Ultra is overkill for anyone not actively using Veo video generation. 9to5Google flagged last week that Google is preparing an Ultra Lite tier to fill the gap[1].

Here is what is plausible and who should buy in.

The full consumer AI pricing landscape

Consumer AI subscriptions, May 2026
SpecMonthly priceBest for
Gemini free$0Casual queries, limited usage
Google AI Pro$20Light Gemini use, 2TB storage
Google AI Ultra Lite (new)~$50-80Heavy Gemini, no compute-intensive Veo
Google AI Ultra$250Power users — Veo, full agentic access
ChatGPT Plus$20GPT-5.5 Instant + Operator
ChatGPT Pro$200GPT-5.5 advanced + extended limits
Claude Pro$20Claude Opus 4.7 with daily limits
Claude Max$100-200Higher Opus 4.7 limits + agents

Pricing has become genuinely confusing. The simplest way to think about it:

  • $0-20/month: Casual use. One vendor. Will hit caps if you use AI for real work.
  • $50-100/month: Heavy single-vendor use. Power user.
  • $200+/month: Multi-modal heavy use (video gen, agent use, long-context document work) or multi-vendor stack.

Ultra Lite is plugging the $50-100 hole on Google's side.

Who is the target

Three personas Ultra Lite is built for:

  1. The heavy Gemini user who does not need Veo. Video generation is the most compute-heavy feature in the Ultra bundle. If you don't use it, the $250 Ultra price is overkill for what amounts to "raise my Gemini quotas". Ultra Lite raises quotas without the Veo budget.
  2. The Workspace power user. Tight integration with Gmail, Calendar, Drive, Sheets makes Gemini the natural choice for Workspace-heavy work. Ultra Lite is the price point where you can use Gemini heavily across the day without anxiety about hitting limits.
  3. The Android power user. Android 17's Gemini-system-service architecture means heavy Android use translates to heavy Gemini use. Ultra Lite normalises the cost for these users.

Who Ultra Lite is not for: developers building products on Gemini's API (use the API directly), or users who genuinely need Veo (just buy Ultra).

How does it compare to a stacked subscription?

The most common alternative right now is to stack ChatGPT Plus ($20) + Claude Pro ($20) + Gemini free. That gets you three frontier models at the consumer tier for $40/month.

If Ultra Lite lands at $50, the stacked option is still cheaper. But it has three downsides:

  • You manage three accounts.
  • You hit caps on each at different times.
  • You don't get the Workspace integration that makes Gemini special.

For a Workspace-heavy user, Ultra Lite probably wins at $50. At $80, it is a coin flip.

What pricing band is realistic

Three signals to read:

  • $50 is the band that makes Ultra Lite a genuine competitor to ChatGPT Plus + Claude Pro stacked. Aggressive but doable.
  • $60-70 is the band that prices Ultra Lite as "more than Plus, less than Pro" — the most likely landing zone based on Google's history of premium pricing.
  • $80+ prices Ultra Lite into "almost Ultra without the perks". This would be a strategic mistake; expect Google to avoid it.

My bet: $60. Slightly above ChatGPT Plus stacked but with the integration moat Google has on Workspace and Android.

What is bundled (educated guesses)

Based on 9to5Google's reporting[1] and Google's history with tiered AI offerings, Ultra Lite probably includes:

  • Higher Gemini quotas across the day — probably 10x the Pro daily quota, vs Ultra's effectively unlimited.
  • Gemini 4 Pro access (the new tier of model expected at I/O).
  • No Veo video generation — this is the feature being held back to differentiate Ultra.
  • Limited Remy / agentic access — probably enough to evaluate, not enough for power use.
  • 2TB Drive storage, same as Pro.

The clean differentiation: Pro is light use, Ultra Lite is heavy Gemini use, Ultra adds Veo and full agentic.

What I would do

Two scenarios:

  1. If you are deep in Google Workspace. Wait for the announcement (probably at I/O), then move to Ultra Lite. It will save you the headache of managing multiple AI accounts and unlock Workspace AI features you cannot match elsewhere.
  2. If you are a multi-vendor power user. Stay on the stacked Plus + Claude Pro + Gemini free setup. Ultra Lite does not give you Opus 4.7 or GPT-5.5 Instant; you would still need the other subscriptions.

For developers building on Gemini specifically — Ultra Lite is consumer pricing, not API pricing. Use the API and pay per-token; the consumer plans are not optimised for programmatic use.

The verdict

Ultra Lite is a smart move. The $20 Pro tier was undersized for real users; the $250 Ultra tier was oversized for non-Veo users. A middle tier was overdue.

The question is execution. If it lands at $50 with Gemini 4 Pro included, it is the consumer AI deal of the year. If it lands at $80 with the same features, it is a polite shrug. Watch the I/O keynote on May 19 for the verdict.

References

  1. [1]

    Google AI Ultra Lite plan for Gemini9to5Google

    https://9to5google.com/2026/05/05/google-ai-ultra-lite-gemini-usage-limits/

Comments

Sign in to leave a comment.

By signing in, Sarma will receive your name, avatar, email, sign-in provider, and approximate location (country/city, derived from your IP) for moderation and reply purposes. None of this is shown publicly — only your name and avatar appear on the post. No newsletter, no marketing, no third-party sharing.

Loading comments…
S

Sarma

Independent software engineer — AI systems, automation platforms, and modern infrastructure.

More in AI

Work with Sarma

Have a project in mind?

I take on a small number of projects each quarter — AI systems, automation, infrastructure, and full-stack engineering.

Get in touch