Oriel — AI & technology worth noticing, every day.

Gemini 4 Carbon

Web page · https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-carbon

First seen by Oriel on .

AI summary

Google introduces Gemini 4 Carbon, a frontier model focused on long-horizon agentic work and cybersecurity, reporting top benchmark results alongside improved alignment and misuse resistance.

  • Carbon ranks #1 on the Artificial Analysis Intelligence Index v5.0 with a score of 59, and places #4 on WebDev Arena at 1769 after an 89-point gain.
  • It scores 88.2% on SEC-bench Pro for vulnerability discovery, 54.8% on ExploitGym for exploitation, and 71.0% on CWE-bench for patching.
  • Google reports more than 2x Argon's success rate for generating secure code; the Chrome team reported over 10X better patch generation than another lab's large frontier model.
  • Safety results include 0.0% successful cheating on Terminal-Bench 4.0, nearly 80% fewer high-severity misalignment alerts, and 0.0% attack success on Gray Swan's Indirect Prompt Injection benchmark.
  • Access begins with selected cyber defenders and expands to trusted partners through the Fairwind Program; the introductory price matches Argon at $2 per million input tokens and $10 per million output tokens.

#gemini #cybersecurity #agentic ai #model safety

Generated by gpt-5.6-sol. Check the original for details.

Metric trends

Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained

Points

Score reported by the source.

  • Hacker News: 1,009, latest observation ; +920 over observed period since

Comments

Comment count reported by the source.

  • Hacker News: 800, latest observation ; +763 over observed period since

Aliases

SourceExternal IDURL
Hacker News50276792https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-carbon/ Read discussion