Google reveals new Gemini 4 Carbon AI model, narrowly beating Claude Opus 5.5 — frontier AI model is also the least likely to hallucinate, benchmarks claim
Last week, Google released its latest AI model, Gemini 4 Carbon, topping intelligence rankings charts.
Google has unveiled its latest AI model, Gemini 4 Carbon, which has quickly claimed the top spot on major intelligence rankings. The model outperforms Claude Opus 5.5 by just one point on the latest Intelligence Index V5 benchmark suite, and excels across four out of nine evaluations. However, what truly sets Gemini 4 Carbon apart is its impressive ability to minimize hallucinations and provide accurate information recall, ranking highest on Artificial Analysis' new AA-Omniscience v1.1 test.
Despite being more expensive than some competitors, its cost per task is slightly lower than Gemini 3.8 Flash and on par with other top models. Google is offering early access to cyber defenders and plans to expand availability through its Fairwind program. This move signals a more cautious approach to model release, with rigorous testing and safety measures in place.
Written by urgent.news from Tom's Hardware's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
- After testing it on 100 photos, Google Photos' new Gemini integration is both good and horrible androidpolice.com
- Google has a growing Gemini problem that readers find hard to ignore androidauthority.com
- Local Government Elections: I turned to Google Gemini for advice, and this is what it told me iol.co.za
