Google's Gemini 4 Carbon leads several benchmarks against GPT-6 Astra and Claude Opus 5.5

AIRohan Paul relays Google DeepMind's announcement of Gemini 4 Carbon, a new frontier model that Google says is state-of-the-art in cybersecurity and long-horizon work. Carbon scores 88.2% on SEC-bench Pro, 54.8% on ExploitGym, 71.0% on CWE-bench, and 96.2% on GraphWalks, against GPT-6 Astra's 71.8% and Claude Opus 5.5's 66.8% on the long-context test.

AIWhy it matters.

The post gives per-benchmark scores against Gemini 4 Argon, GPT-6 Astra, and Claude Opus 5.5, so readers can compare cybersecurity and long-context results directly.

Rohan Paul

@rohanpaul_ai

View on X

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%.

Image from @GoogleDeepMind's post — Google DeepMind@GoogleDeepMind

Source details

Rohan Paul · x.comPublished Oct 28, 1:28 PM ET · added here 1:39 PM ET

Original title: Gemini 4 Carbon, Google's new frontier model, beats GPT-6 Astra and Claude Opus 5.5 on Artificial Analysis.

Google's Gemini 4 Carbon leads several benchmarks against GPT-6 Astra and Claude Opus 5.5

AIWhy it matters.

Topics