Google announces Gemini 4 Carbon, released first to a subset of cyber defenders
Google said its new frontier model leads several cybersecurity benchmarks and will reach a subset of cyber defenders this week before developers, businesses and consumers.
AI illustrationGoogle has announced Gemini 4 Carbon, which it describes as its strongest model, with state-of-the-art cybersecurity and long-horizon capabilities. The model builds on Gemini 4 Argon, which Google said was launched two weeks ago.
Access will be staged. Google said that starting this week it will make Carbon available to a subset of cyber defenders and will gradually phase access to additional trusted partners through its Fairwind Program. The company said it is engaged in the U.S. government's voluntary process for pre-release model testing and will keep gathering feedback and iterating on guardrails before offering the model to developers, businesses and consumers via its product surfaces as soon as possible.
Google said Carbon ranks #1 at 59 on the recently updated v5.0 of the Artificial Analysis Intelligence Index. On external cybersecurity benchmarks, evaluated without guardrails, it reported 88.2% on SEC-bench Pro for finding vulnerabilities, 54.8% on ExploitGym for exploitation and 71.0% on CWE-bench for patching. In web development, Google said the model gained 89 elo points to reach #4 (1769) on WebDev Arena.
On internal benchmarks, Google said Carbon achieves more than 2x uplift in success rate for generating secure code over Argon. Its Chrome team reported that Carbon performed more than 10X better than another lab's large frontier model when generating high quality patches. Google also said its teams have seen Carbon run for days on complex, multi-step engineering tasks such as root-causing and patching subtle bugs across large codebases.
On alignment, Google said successful cheating on Terminal-Bench 4.0 fell to 0.0%, with attempts to cheat dropping to 0.4%, and that high-severity misalignment alerts dropped by almost 80% in live coding environments with its internal teams. In internal testing, Google said Carbon cuts unsupported claims by approximately 2x over Argon and reduced overreach by 26% and under-service by 29%. On the Gray Swan Indirect Prompt Injection Benchmark, Google said Carbon achieves a 0.0% attack success rate, against 0.7% for Argon.
Google said Carbon will launch at the same introductory price as Argon: $2 per million input tokens and $10 per million output tokens, with cached input tokens priced at 95% off the input token price.