Z.ai launches GLM-5.2 for long-horizon tasks
Chinese AI company Z.ai has launched GLM-5.2, a new flagship model optimised for complex tasks and agentic coding.
What happened?
Z.ai (Zhipu AI) has introduced GLM-5.2, a new large language model (LLM). The model is specifically designed to handle "long-horizon tasks", which include complex operations such as agentic coding and software engineering. GLM-5.2 is built on a Mixture-of-Experts (MoE) architecture with 744 billion parameters.
Key facts
| Modellnamn | GLM-5.2 |
|---|---|
| Lanseringsdatum (Hugging Face) | 25 juni 2026 |
| Arkitektur | Mixture-of-Experts (MoE) |
| Antal parametrar | 744 miljarder |
| Huvudsaklig optimering | Long-horizon tasks |
”We're introducing GLM-5.2, our latest flagship model for long-horizon tasks. It marks a substantial leap in long-horizon task capability over its predecessor...”
”GLM-5.2 is Z.ai's newest flagship model for agentic coding and long-horizon software engineering, built on the GLM-5 Mixture-of-Experts backbone — 744B...”
”The NVIDIA GLM-5.2 NVFP4 model is the quantized version of ZAI's GLM-5.2 model, commercial or non-commercial use. Release Date: Hugging Face 06/25/2026”
Why it matters
The launch of GLM-5.2 indicates the continued development of AI models for specialised applications. The focus on "long-horizon tasks" and agentic coding could improve automation and efficiency within software development and other complex problem areas. The model positions itself as a competitor in the field of advanced problem-solving capabilities.
Who is affected?
Developers and companies working with agentic systems, software engineering, and complex information management tasks are directly affected. These users can benefit from GLM-5.2's capacity to handle large and interconnected tasks. Users reliant on AI assistants for coding and problem-solving may also see improvements in quality and efficiency.
Impact on the EU
GLM-5.2 is available via API and as a quantised version on Hugging Face. The model can be used both commercially and non-commercially. Although Z.ai is a Chinese company, global availability via platforms such as Together.ai makes it accessible to developers and companies within the EU.
What else you should know
A quantised version of GLM-5.2, named NVIDIA GLM-5.2 NVFP4, was released on Hugging Face on 25 June 2026. Information regarding the model initially circulated on Reddit, but its official launch was confirmed via Hugging Face and Together.ai.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vilka bolag berörs?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Decide whether this affects strategy over 6–12 months or is just noise.
- Discuss with leadership: do we own the right question or does ownership need to move?
- Ask: what risk are we taking by NOT acting on this this quarter?
Generated angle — not editorial analysis of "Z.ai launches GLM-5.2 for long-horizon tasks"