unsloth / text-generation updated 1 month ago

GLM-5.2-GGUF

Name: GLM-5.2-GGUF
Author: unsloth

See Unsloth Dynamic 2.0 GGUFs for our quantization benchmarks. You can now run GLM-5.2 in Unsloth Studio with toggles for High and Max thinking. Read our GLM-5.2 guide for analysis and instructions. See below for example of 1-bit UD-IQ1M GGUF running in Unsloth:

Params

—

Context

—

Downloads 30d

238 K

Likes

618

Commercial use: allowed mit Not gated GGUF 2 languages View on Hugging Face ↗

Download history

daily snapshots · 10 days

344 K238 K

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
GLM-5.2-BF16-00005-of-00033.gguf	GGUF	46.8 GB	52.0 GB	❌ Won’t fit
GLM-5.2-UD-IQ2_M-00002-of-00006.gguf	IQ2_M	49.2 GB	54.6 GB	❌ Won’t fit
GLM-5.2-UD-IQ2_XXS-00003-of-00006.gguf	IQ2_XXS	49.1 GB	54.6 GB	❌ Won’t fit
GLM-5.2-Q8_0-00003-of-00017.gguf	Q8	49.3 GB	54.7 GB	❌ Won’t fit
GLM-5.2-UD-IQ1_S-00003-of-00006.gguf	IQ1_S	49.7 GB	55.2 GB	❌ Won’t fit
GLM-5.2-UD-IQ1_M-00005-of-00006.gguf	IQ1_M	49.9 GB	55.4 GB	❌ Won’t fit
GLM-5.2-UD-IQ3_S-00003-of-00008.gguf	IQ3_S	49.9 GB	55.4 GB	❌ Won’t fit
GLM-5.2-UD-IQ3_XXS-00002-of-00007.gguf	IQ3_XXS	50.0 GB	55.5 GB	❌ Won’t fit

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub

~ · ollama · GGUF

$ ollama run glm-5-2-gguf

# pin the quantization explicitly
$ ollama run glm-5-2-gguf-gguf

est. VRAM 52.0 GBon RTX 4090 · 24 GBJSON API →

Specifications

Licence: mit
First seen on the Hub: 2026-06-17
Base model: GLM-5.2
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Family

Base model and the most-downloaded derivatives in the catalog.

Base GLM-5.2 zai-org 2.2 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB