MODEL FAMILY

GLM 5.2

Choose the size and format you want to run. Compare each version’s hardware fit, model scores, and setup guides in one place.

TextCodingAgents

2versions to explore

1published guide

Compare versions

PERSONALIZE THE COMPARISONWhich machine will you use?

Save your hardware to compare which versions can run on your machine.

Add my hardware

FIND YOUR VERSION

Compare versions

Scores and requirements stay with each version.

753.33BText

GLM 5.2

GLM-5.2 is an MIT-licensed text-generation model from zai-org, presented as a flagship model for long-horizon tasks with a 1M-token context. It is tagged with safetensors and text-generation, supports English and Chinese, and uses the transformers library. The card describes improved architecture, coding capabilities with multiple thinking effort levels, and local serving support via SGLang, vLLM, Transformers, and KTransformers, with a newer version GLM-5.3-BF16 noted.

Model size: 753.33B
Format: FP16 weights
Context: 1,048,576

Freshness 63 out of 100. Repository age: 121 days old, Jun 16, 2026. Measures age, not quality.

Popularity 80 out of 100. 492,903 downloads in 30 days; 5,181 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.

SETUP GUIDES FOR THIS VERSIONBuild a coding assistant with GLM 5.2 on rented GPUs

Explore this version Model source

753.33B · FP8Text

GLM 5.2 FP8

GLM 5.2 FP8 is an FP8 quantized text-generation model from zai-org, released under the MIT license. Its official model card describes it as a flagship model for long-horizon tasks with a solid 1M-token context, improved coding, an IndexShare architecture, and MTP speculative decoding. The card lists SGLang, vLLM, Transformers, and KTransformers as deployment frameworks, and links to blog, paper, API, and community resources.

Model size: 753.33B
Format: unknown weights
Context: 1,048,576

Freshness 63 out of 100. Repository age: 121 days old, Jun 16, 2026. Measures age, not quality.

Popularity 70 out of 100. 196,139 downloads in 30 days; 268 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.

SETUP GUIDES FOR THIS VERSIONBuild a coding assistant with GLM 5.2 on rented GPUs

Explore this version Model source