Google’s Gemini 4 Carbon now leads the Artificial Analysis Intelligence Index v5. It scores 59 on the index. It tops four of nine index evaluations. It ranks second on the Coding Agent Index behind Claude Opus 5.5. It costs 23% less per task than Claude Opus 5.5. It uses 29% fewer output tokens per task than Claude Opus 5.5. It has a 1M token context window. It supports text, image, video and speech input. It offers standard pricing of $4/$20 per 1M input/output tokens. It is not yet publicly available. Google offers a 50% launch discount with no confirmed end date.
What changed
Gemini 4 Argon was the previous top-scoring model on Intelligence Index v5.
What it unlocks
Access to a model with leading performance in complex reasoning, knowledge, and scientific tasks at a lower cost per task than the prior leader.
- 59 Intelligence Index v5 score
- $7.76 per Intelligence Index task
- 123k output tokens per task
- $4/$20 per 1M input/output tokens
What you need to act on it
- selected user access
- 50% launch discount period
Sources