Google has begun releasing its Gemini 4 Argon model to vetted cyber defenders, with paid API customers and AI Ultra subscribers to follow in a phased rollout, with no dates announced.
Google said the model is “rolling out to a set of trusted cyber defenders through our Fairwind Program” and is also in use inside the company. Koray Kavukcuoglu, senior vice-president of Google DeepMind and chief AI architect at Google, wrote the 30 September announcement.
Google said wider access will run in this order: developers, enterprises, then consumers, “starting with paid API customers and Google AI Ultra subscribers”. It said it is taking part in the US government’s voluntary pre-release access process. As of early October, the Gemini app and public API did not list the model.
Google describes Argon as built for software engineering, enterprise work such as legal and finance tasks, and cyber defence. It reported these results, which have not been independently verified:
9% on DeepSWE v1.1, a software engineering benchmark.
3% on Zapier’s AutomationBench, first place.
7% on LVBench, which tests long-video understanding.
68% on CWE-bench v1, a vulnerability remediation test, tied for first.
Google also said the maximum output length rises to 1 million tokens, from 64,000.
Introductory API pricing is $2 per million input tokens and $10 per million output tokens. It rises to $4 and $20 after the introductory period. Cached input is priced at 95% below the input rate.
On security, Google said it will release Argon without cyber guardrails to trusted defenders and internal teams. For other users, it said the model refuses harmful cyber and chemical, biological, radiological and nuclear requests. It said it monitors the model’s reasoning and actions and halts execution when needed.
Google said its agents are migrating C and C++ code to Rust, including more than 800,000 lines in the Fuchsia Zircon kernel. It said security firm Wiz is using Argon through its Scan for Good initiative and found a critical vulnerability in healthcare software that earlier models missed.
For technology leaders, the practical point is timing. Google has set no date for the paid API or Ultra tiers, and the headline pricing is introductory.
Click Here For The Original Source
——————————————————–
