NewsGoogle2 min read
Google Unveils Gemini 4 Argon With a 1M-Token Output Limit
Google announces Gemini 4 Argon, its new frontier model for coding, professional work and cybersecurity. Pricing, availability and what changes.
Key points
- Announced on 30 September 2026 as Google's new frontier model.
- It can generate up to 1 million output tokens, up from 64,000 in the previous generation.
- Launch price: $2 per million input tokens and $10 per million output; later $4 and $20.
- For now only selected cybersecurity defenders can use it; the paid API and Google AI Ultra come next.
⚠ Pending verification. We have not yet confirmed these details with an official source. Check them before relying on them:
- How long the launch price lasts: Google has not given a date.
- When it reaches the paid API and Google AI Ultra: Google only says “as soon as possible”.
Google has unveiled Gemini 4 Argon, the model it describes as its “next era of frontier intelligence”. Koray Kavukcuoglu of Google DeepMind announced it on 30 September. It is built for long tasks that need many steps of reasoning: real software projects, professional work in fields such as law and finance, and cybersecurity defence.
What was announced
The headline change is the output limit. Gemini 4 Argon can write up to 1 million tokens in a single response, where the previous generation stopped at 64,000. This is not the context window, which is how much the model can read; it is how much it can produce in one go, such as a very long report or an entire codebase.
Google published its own benchmark results alongside the announcement:
| Benchmark | What it measures | Result according to Google |
|---|---|---|
| DeepSWE v1.1 | Real-world software engineering | 77.9% |
| AutomationBench | Task automation | 51.3% (first place) |
| CWE-bench v1 | Fixing vulnerabilities | 68% (tied for first) |
| LVBench | Understanding long videos | 91.7% |
These are the company’s own figures. It is worth waiting for independent evaluators to reproduce them; our benchmark entry explains why.
Price and availability
The model launches at an introductory price of $2 per million input tokens and $10 per million output tokens. After that period it rises to $4 and $20. Input tokens reused through prompt caching are 95% cheaper.
It is not open to the public yet. Google is giving it first to a group of trusted cyber defenders through its Fairwind Program, and has offered voluntary early access to the US government. Paid API customers and Google AI Ultra subscribers come next.
What changes for you
- If you use Gemini day to day, nothing yet: you will have to wait until it reaches the app and your plan.
- If you build with the API, the output limit opens up jobs that used to need splitting across many calls. Watch the cost: 1 million output tokens is $10 at the launch price.
- If you compare models, the launch price puts it level with mid-range models. We will add it to our model comparison once it is generally available.
What was said on X
The official announcements, as they were posted. Tap to open them on X.
- Logan Kilpatrick@OfficialLoganKIntroducing Gemini 4 Argon, our new frontier model, rolling out to cyber defenders starting today, and more widely as soon as possible. I am really excited by the progress we have made here. Argon is priced at $2 in and $10 out during introductory pricing!View on X
- Google@GoogleGemini 4 Argon is our next era of frontier intelligence. It shows significant improvements across benchmarks, setting a new state of the art for real-world long-horizon software engineering tasks.View on X
