GLM-5.2, the latest flagship from Z.ai (Zhipu AI), has ranked second in the Game Dev Arena category of Design Arena—a benchmark that ranks the quality of AI-generated content via user votes—with an Elo of 1368. That marks a jump of six positions and 29 Elo points over the previous GLM-5.1, putting it in the same performance band as Anthropic's Claude Fable 5. It is the top open-weight model in the category and ranks second among labs, behind only Anthropic.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.