Google has announced Gemini 4 Carbon, its next frontier AI model, following the launch of Gemini 4 Argon just two weeks ago. According to Google, Carbon is the company's strongest model yet, delivering state-of-the-art cybersecurity and long-horizon capabilities. The announcement was made by Koray Kavukcuoglu, SVP of Google DeepMind and Chief AI Architect.
Carbon ranks number one on the recently updated v5.0 of the Artificial Analysis Intelligence Index with a score of 59. This rapid release cycle illustrates the pace at which Google is pushing the frontier forward, with researchers and models working hand in hand.
Gemini 4 Carbon benchmarks and capabilities
Google reports significant gains across coding, agentic tasks, and key capabilities critical for cybersecurity defense. The model shows a firm lead in several specialized benchmarks.
On internal benchmarks, Carbon drives a strong uplift in generating secure code, and in live coding environments with Google's own teams, it shows reduced misalignment and fabrication rates. Based on user feedback following the last launch, Google worked to improve Carbon across coding, agentic, and scientific tasks, as well as full-stack web development and multimodal understanding.
- Artificial Analysis Intelligence Index v5.0: 59 (ranked #1)
- SEC-bench Pro (finding vulnerabilities): 88.2%
- ExploitGym (exploitation): 54.8%
- CWE-bench (patching): 71.0%
