Gemini 4 Carbon tops GPT-6 Astra and Opus 5.5 on Artificial Analysis
Gemini 4 Carbon leads on security benchmarks (88.2% SEC-bench Pro, 71.0% CWE-bench), rolls out first to cyber defenders at $2/$10 per 1M tokens, and scores 96.2% on GraphWalks long context.
Gemini 4 Carbon, Google's new frontier model, beats GPT-6 Astra and Claude Opus 5.5 on Artificial Analysis. > Arriving 2 weeks after Gemini 4 Argon, Carbon gains most in security: 88.2% on SEC-bench Pro for finding vulnerabilities and 71.0% on CWE-bench for patching. 54.8% on ExploitGym for exploiting them and 71.0% on CWE-bench for patching them. > Because of that exploit skill, access starts this week with a subset of cyber defenders, then trusted partners through the Fairwind Program, before developers, businesses and consumers get it. > Google says Carbon can work on engineering problems for days, root-causing subtle bugs across large codebases instead of stalling at dead ends. > Cost: introductory $2/$10 per 1M input/output tokens, later doubling to $4 and $20. > On long-context benchmark: on GraphWalks (256K to 1M tokens, BFS F1), Carbon scores 96.2%, against Argon’s 84.2%, GPT-6 Astra’s 71.8% and Claude Opus 5.5’s 66.8%.
This post quotes another post: “Google DeepMind unveils Gemini 4 Carbon as next frontier model”