Zhipu GLM-5.1 High-Speed released, setting a new global record for LLM API speed
Original title:智谱 GLM-5.1 高速版 AI 模型发布,跑出全球最快速度 400 tokens/s
AI summary
On May 22, Zhipu released a high-speed API version of its flagship large model GLM-5.1, "GLM-5.1-highspeed," to select enterprise customers. The version delivers output speeds of 400 tokens/s, setting a new global ceiling for LLM API speed. The key breakthrough is that it is the first domestic large model to combine flagship-level capability with low latency, breaking the convention that a high-speed model is a lightweight model. The version was developed by the Zhipu GLM team in collaboration with the TileRT team, with system-level optimizations ensuring production-grade stability of the speed, suited to scenarios such as AI coding and real-time voice interaction.
Key Takeaways
- On May 22,Zhipu released a high-speed API version of its flagship large model GLM-5.
- 1-highspeed," to select enterprise customers.
- The version delivers output speeds of 400 tokens/s,setting a new global ceiling for LLM API speed.
- The key breakthrough is that it is the first domestic large model to combine flagship-level…
- The version was developed by the Zhipu GLM team in collaboration with the TileRT team,with…
Quick Facts
- Published
- Updated
- Source
- IT之家·人工智能
- Category
- Model Releases
- AI score
- 73
智谱把旗舰模型拉到 400 tokens/s,还保持全尺寸能力,不是那种为快阉割的小模型。做实时交互、AI 编程的可以认真看看,延迟敏感场景的选型参数要重写了。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Zhipu GLM-5.1 High-Speed released, setting a new global record for LLM API speed. May 22, 2026. https://new.nat6.net/en/news/cmpga762r0dm4sljw23zihurw/