Gemini 4 Carbon edges out Claude Opus 5.5 by one point in Artificial Analysis ranking
Google's new model, Gemini 4 Carbon, has taken first place in the AI model ranking run by Artificial Analysis, Tom's Hardware reports. It scored 59 points, one more than Claude Opus 5.5 and two more than GPT-6 Astra. It is also cheaper than Opus: it completes a single test task for $7.76, compared with $10.04 for Opus. It also came out on top in a test that measures how much a model knows and how often it makes things up. For now, Google is making it available only to select cybersecurity teams.