Open-Weight Model Families · 2025
GLM-4.5 to GLM-5.2 (Zhipu AI / Z.ai open-weight MoE series)
Put a frontier line of large mixture-of-experts models into the open-weight ecosystem: GLM-4.5 through the GLM-5.2 flagship, released by Zhipu AI / Z.ai under an MIT license with downloadable weights rather than API-only access.
Editorial record
Plain-language summary
From GLM-4.5 (a 355B-parameter sparse mixture-of-experts model with 32B active, July 2025) through the GLM-5 family and the GLM-5.2 flagship (a roughly 750B-parameter MoE with sparse attention and a million-token context, 2026), Zhipu AI released the weights on Hugging Face under MIT, with FP8 variants for cheaper serving. The series targets reasoning and agentic software engineering and is also served through the z.ai API, but the checkpoints are genuinely public and self-hostable. It is among the most capable open-weight lines outside the Llama, Qwen, and DeepSeek families. (GLM-5.x release dates are approximate, pinned from repository timestamps rather than day-precise announcements.)
Knowledge graph
Relationships
Descendants
Conceptual ancestorEvidence: Strongly supported
GLM: General Language Model Pretraining with Autoregressive Blank Infilling / GLM-130B
The modern open-weight GLM series continues the GLM lineage
zai-org HF
Depends onEvidence: Strongly supported
DeepSeek (V2/V3; MLA + efficient MoE + FP8)
GLM-5 adopts DeepSeek-style sparse attention
GLM-5 report 2602.15763
Depends onEvidence: Strongly supported
Switch Transformer
GLM-4.5 and 5.x are large sparse mixture-of-experts models
GLM-4.5 report 2508.06471
Source record
Provenance
- Record ID
- P-367
- Record created
- 2026-07-13
- Last reviewed
- 2026-07-14
- Record version
- 2
- https://arxiv.org/abs/2508.06471
- arXiv:2508.06471
Citation caveat: Citation metadata is approximate and marked unverified in the source dataset.