Google's Gemini 4 Carbon is the new number one on the Artificial Analysis Intelligence Index v5, scoring 59 points. The model, announced on October 28, 2026, leads four of the nine evaluations that make up the index and is being rolled out to selected users rather than the general public.
The result matters because it reshapes the top of the AI leaderboard. Gemini 4 Carbon also ranks second on the Artificial Analysis Coding Agent Index behind Claude Opus 5.5, a notable improvement for Google, which has historically underperformed in coding agent tasks.
Gemini 4 Carbon tops Artificial Analysis Intelligence Index v5
Gemini 4 Carbon's 59 points are driven by leading performance in complex reasoning, knowledge and scientific tasks. It took the top spot in HLE-Diamond, Terminal-Bench-Science, AA-Omniscience and AutomationBench-AA.
On the recently updated AA-Omniscience v1.1, the model scored 54, six points ahead of Claude Opus 5.5 (max) and eight points ahead of Gemini 4 Argon. Accuracy rose 10 points over Argon to 58%, while its hallucination rate increased from 4% to 11%, still the second lowest of any model behind Argon.
- Intelligence Index v5 score: 59 (high reasoning)
- Leads 4 of 9 evaluations: HLE-Diamond, Terminal-Bench-Science, AA-Omniscience, AutomationBench-AA
- AA-Omniscience v1.1: 54, ahead of Claude Opus 5.5 (48) and Gemini 4 Argon (46)
- Coding Agent Index: 59 with Antigravity CLI, 3 points behind Claude Opus 5.5 (62) in Claude Code
