Gemini 4 Carbon is Google's new frontier model, pitched at long-running agent work and security. Google says it ranks first on the Artificial Analysis Intelligence Index v5.0 at 59, scores 88.2% on SEC-bench Pro (finding vulnerabilities), 54.8% on ExploitGym and 71.0% on CWE-bench (patching), and climbed 89 Elo points to fourth on WebDev Arena (1768). Internally, Google says, Carbon has run for days on multi-step engineering tasks such as root-causing bugs across large codebases.
Google also reports alignment numbers: successful cheating on Terminal-Bench 4.0 down to 0.0% (attempts at 0.4%), high-severity misalignment alerts in live internal coding down by almost 80%, and a 0.0% attack success rate on Gray Swan's indirect prompt-injection benchmark, against 0.7% for Argon.
You can't use it yet. Starting this week Google gives Carbon to a subset of cyber defenders, then more partners through its Fairwind Program, and says developers, businesses and consumers come after that. The introductory price is $2 per million input tokens and $10 per million output, with cached input 95% off. After the intro period it goes to $4 and $20.