Models · scorecard

Gemini 4 carbon

What the vendor claims, next to what independent boards measured. The two are never mixed in one column.

At a glance

OpenRouter does not list this model, so there is no price or context window here.

What the vendor claims

Quoted word for word from the vendor's own launch post or model card. Claims, not measurements.

Scores the vendor claims, each with its source and the quoted line
Benchmark	Claimed	Source
SEC-bench ProSetting: evaluated without guardrailsThe source says: Carbon achieves a state-of-the-art 88.2% on SEC-bench Pro, autonomously uncovering complex flaws buried in real-world software.	88.2%	Google DeepMind · Oct 28 From Drop #053 · Thu, Oct 29, 2026 · Morning
ExploitGymSetting: evaluated without guardrailsThe source says: Carbon sets a new state-of-the-art on the popular exploitation benchmark ExploitGym (54.8%).	54.8%	Google DeepMind · Oct 28 From Drop #053 · Thu, Oct 29, 2026 · Morning

Independent results

Measured by independent sources (Epoch AI, LMArena, OpenRouter's benchmark runs, τ²-bench, Vectara), not by the vendor. "Listed as" is the source's own name for the model, so you can check our match.

No independent results yet. They appear here as Epoch AI and LMArena add this model.

In the Drops

ModelsOct 29, 2026

Google DeepMind announces Gemini 4 Carbon, starting with cyber defenders

Google DeepMind says Gemini 4 Carbon scores 88.2% on SEC-bench Pro and resisted every evaluated prompt-injection attack. Access starts with a subset of cyber defenders; the API will cost $2/$10 per million tokens.

Why it matters: A new frontier option for security and long-running coding work, but builders can't use it yet. Ars notes Google gave no date for a wide release.

Google DeepMind · Oct 28 Ars Technica · Oct 28 Hacker News · Oct 28

From Drop #053 · Thu, Oct 29, 2026 · Morning