Gemini 4 Carbon Now Available: Who Can Use It, Price and How to Try
Google has opened Gemini 4 Carbon, its most powerful AI model, to paying developers, businesses and Google AI Ultra subscribers. Here is who can use it, what it costs and how it compares with Argon.
Image: TeQBlogs
Gemini 4 Carbon, Google’s most powerful AI model, is now open to more people. On 10 November 2026 Google made it available to paying developers and businesses through the Gemini API, and began rolling it out to Google AI Ultra subscribers in the Gemini app, AI Mode in Search and the Antigravity coding tool. Developers get an introductory price of $2 per million input tokens and $10 per million output tokens until 30 November 2026, after which it doubles.
Quick summary
- What it is: Google’s “next frontier model”, built for long, multi-step tasks, coding and cybersecurity.
- First released: 28 October 2026, only to a small group of cyber defenders.
- Now (10 November 2026): paid Gemini API users, Gemini for business, and Google AI Ultra subscribers.
- API price: $2 input / $10 output per million tokens until 30 November 2026, then $4 / $20.
- Not for everyone yet: free Gemini users and other Google AI plans are not part of this rollout.
- Google’s advice: Gemini 4 Argon is still the better choice for most everyday coding and office work.
What is Gemini 4 Carbon?
Gemini 4 Carbon is the second model in Google’s Gemini 4 family, following Gemini 4 Argon. Google DeepMind calls it its strongest model, with state-of-the-art cybersecurity and “long horizon” abilities. In practice, that means it can keep working on a hard task for a long time, try different approaches when it hits a dead end, and check its own work.
Google says its internal teams saw Carbon run for days on multi-step jobs such as finding the root cause of a bug and patching it. It also improves on coding, science tasks, full-stack web development and understanding images and other media. The same family of models powered the AI agents in DeepMind’s recent AlphaProof Code verification of Signal’s encryption.
Also read: CERT-In AI Security Auditing Guidelines: Risk Tiers, Audit Phases and Checklist
Who can use Gemini 4 Carbon now?
Access depends on who you are. This table sums up Google’s 10 November announcement.
| User type | Where to use Carbon | Status |
|---|---|---|
| Google AI Ultra subscribers | Gemini app, AI Mode in Search, Antigravity, AI Studio | Rolling out |
| Developers | Gemini API paid tier in AI Studio and Android Studio | Available |
| Businesses | Gemini for business (API or model drop-down), Antigravity, CodeMender | Available |
| Scientists | Free access to Argon and Carbon in Antigravity’s top usage tier for 6 months | By application |
| Cybersecurity defenders | “Carbon for cyber” via Google’s Fairwind Program | Rolling, vetted access |
| Free Gemini users and other plans | Not included | No date announced |
Google’s announcement does not list individual countries. If you pay for Google AI Ultra in India, open the Gemini app and check the model picker; because this is a gradual rollout, it may take a few days to appear.
Gemini 4 Carbon API pricing
Developers pay per token, and the price will rise after the launch offer ends. Carbon thinks longer than Argon, so the same task can use more tokens and cost more.
| Pricing (per 1 million tokens) | Until 30 November 2026 | From 1 December 2026 |
|---|---|---|
| Input | $2 | $4 |
| Output | $10 | $20 |
| Cached input | 95% off the input price | 95% off the input price |
The introductory price is the same as Gemini 4 Argon’s launch price. Prices are in US dollars, as listed by Google at the time of writing.
How good is Gemini 4 Carbon? Key benchmarks
Google’s published results put Carbon at or near the top of several independent and partner benchmarks:
- #1 on the Artificial Analysis Intelligence Index v5.0, with a score of 59.
- 88.2% on SEC-bench Pro (finding vulnerabilities) and 71.0% on CWE-bench (patching them).
- #1 on CursorBench 4.0, Perplexity’s WANDR and Zapier’s AutomationBench, at more than 2x lower cost per task than the next-best external model, according to Google.
- 92% on a 375-task agentic honesty benchmark, up from 70% for Argon.
These are Google’s own figures. Real results depend on your task, so test Carbon on your own work before switching everything over.
Gemini 4 Carbon vs Gemini 4 Argon: which should you use?
Google itself says Argon “remains the ideal choice for the majority of software engineering and knowledge work”. Use Carbon for the hard stuff:
- Choose Carbon for long agent jobs, complex debugging, security reviews and multi-step research where accuracy matters more than speed.
- Choose Argon for everyday coding, writing, summaries and office tasks, where it is faster and uses fewer tokens.
Google also previewed /swarm in Antigravity. One prompt turns into a team of AI agents: Carbon handles the plan, coordination and review, while lighter models work in parallel and separate “verifier” agents test each change.
How to try Gemini 4 Carbon
- Google AI Ultra users: open the Gemini app or gemini.google.com and pick Gemini 4 Carbon from the model menu once it reaches your account. It also appears in AI Mode in Search.
- Developers: sign in to Google AI Studio on a paid API key and select the model “gemini-4-carbon”.
- Businesses: choose Carbon from the model drop-down in Gemini for business, or call it through the API.
- Watch your token use in the first week, especially after 30 November when the price doubles.
Comparing AI plans? See our guide to ChatGPT plans in India, how to convert Gemini Gems to Skills, and what’s new in the November Pixel Drop.
Frequently asked questions
Is Gemini 4 Carbon free?
No. It is available to Google AI Ultra subscribers and to developers on the paid Gemini API tier. Google has not announced access for free Gemini users.
How much does the Gemini 4 Carbon API cost?
$2 per million input tokens and $10 per million output tokens until 30 November 2026, then $4 and $20. Cached input is 95% cheaper.
What is the difference between Gemini 4 Carbon and Argon?
Carbon is Google’s most capable model for long, difficult tasks and security work, and uses more tokens to explore and verify. Argon is faster and recommended for most everyday coding and knowledge work.
When was Gemini 4 Carbon released?
Google announced it on 28 October 2026 for a small group of cyber defenders, and opened it to paying developers, businesses and AI Ultra subscribers on 10 November 2026.
The bottom line
Gemini 4 Carbon is Google’s answer for the hardest AI jobs: long agent tasks, deep debugging and security work. If you pay for Google AI Ultra or build on the Gemini API, it’s worth testing now while the launch price lasts until 30 November. For everyday chat, writing and most coding, Gemini 4 Argon is still the faster, cheaper pick.
Sources
- Google: Gemini 4 Carbon is now available (10 November 2026)
- Google: Gemini 4 Carbon, our next frontier model (28 October 2026)
About The Author
