Back to zai-org
zai-org /

GL
GLM-5.2

Open WeightsChat & ReasoningtextUpdated June 16, 2026

GLM-5.2

Model Overview

GLM-5.2 is a 753.3B-parameter Mixture-of-Experts text generation model developed by zai-org. Built on the GlmMoeDsaForCausalLM architecture. Supports en, zh. Released on 2026-06-16.


📊 Quick Specs

Specification Table
SpecificationValue
Parameters753.3B
ArchitectureGlmMoeDsaForCausalLM
Tasktext generation
Modalitytext
LicenseMIT
Frameworktransformers
MoEYes
Languagesen, zh

✨ Key Features

  • 753.3B parameters with sparse MoE architecture for efficient inference
  • Built on GlmMoeDsaForCausalLM architecture (transformers)
  • Primary task: text generation (text modality)
  • Language support: en, zh
  • Open-weights under MIT license — self-hostable and fine-tunable
  • High adoption: 667,403 downloads on Hugging Face

📈 Community Adoption

  • 4,422 likes on Hugging Face
  • 667,403 downloads on Hugging Face

🔗 Resources


📜 License & Access

MIT — Open-weights model available for download, fine-tuning, and self-hosted deployment.

Key Features

753.3B parameters with sparse MoE architecture for efficient inference

Feature 01

Built on GlmMoeDsaForCausalLM architecture (transformers)

Feature 02

Primary task: text generation (text modality)

Feature 03

Language support: en, zh

Feature 04

Open-weights under MIT license — self-hostable and fine-tunable

Feature 05

High adoption: 667,403 downloads on Hugging Face

Feature 06

You might also want to compare

Verified Sources

Tags

open-weightstrendinghuggingfacemoetransformerstext-generation

Model Specs

open-weights

Parameters

753.3B

Context Window

unknown

License

MIT

Deployment

self-hostable

Resources & Links

Curator Notes

Auto-ingested from HuggingFace Trending on 2026-07-25. Architecture: GlmMoeDsaForCausalLM. Params: 753.3B. License: MIT.

Compare Specs

Compare parameters, context windows, modalities, and benchmark scores of this model side-by-side with others.

Compare Model