Skip to content
TopicTracker
From HackerNewsView original
TranslationTranslation

GLM-5.2 Is the New Best Open Model

GLM-5.2 has been released as the new top-performing open-weight model, surpassing previous open models on several benchmarks. It achieves competitive results against proprietary models in reasoning, coding, and multilingual tasks.

Background

- **GLM-5.2** is a new open-weight large language model released by **Zhipu AI** (a major Chinese AI company, also known as 智谱AI). It has topped the **Open LLM Leaderboard**, a community benchmark that tracks which models perform best on a range of reasoning and knowledge tasks, knocking previous leaders like Llama-3 from Meta off the top spot. - Because GLM-5.2 is "open" (its weights are publicly released, allowing anyone to download, fine-tune, and run it), this signals a shift in the global AI landscape: Chinese open models are now competing directly with, and in some cases outperforming, Western ones (e.g., Meta's Llama, Mistral's models). - The "open vs. closed" debate is central here. Open models can be inspected and modified, but critics worry about safety risks; closed models (like OpenAI's GPT-4 or Google's Gemini) are restricted via APIs. A Chinese model topping the open leaderboard raises geopolitical stakes in AI development.

Related stories

  • The author tests the new open-weights GLM 5.2 model against Gemini 3 Flash by having them play text adventure games within a fixed budget of $0.15 per attempt. A harness tracks achievements earned in each game to compare the models' performance.