Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5 (artificialanalysis.ai)

Google's Gemini 4 Carbon has secured the top position on the Artificial Analysis Intelligence Index v5, achieving a score of 59. This model excels in complex reasoning, knowledge, and scientific tasks, significantly improving performance in coding and knowledge work compared to previous Gemini iterations. Despite leading in key categories such as HLE-Diamond and AutomationBench-AA, Gemini 4 Carbon still trails behind competitors like Claude Opus 5.5 in coding tasks and overall agentic knowledge work. Notably, its cost per task is approximately 23% lower than that of Claude Opus 5.5, making it an appealing option for developers and businesses. Gemini 4 Carbon distinguishes itself with a context window of 1 million tokens and enhanced multimodality, allowing it to process text, image, video, and speech inputs. Its performance in the AA-Omniscience v1.1 benchmark reveals a low hallucination rate of 11%, one of the best in the market, alongside an accuracy rate improvement of 10 points over its predecessor, Gemini 4 Argon. The model is currently available to select users with a promotional launch discount, indicating Google's intent to capture market share with this innovative release while demonstrating substantial advancements from its earlier models.