vultr/GLM-5.2-GGUF

GLM 5.2, GGUF

GLM 5.2 is an advanced Mixture-of-Experts language model designed for long-horizon reasoning, large-scale coding, and million-token context processing. The model incorporates IndexShare sparse attention to improve long-context efficiency and an enhanced multi-token prediction layer for faster speculative decoding. It is optimized for complex reasoning, extended coding workflows, and sustained long-context tasks.

This repository holds Vultr's GGUF quantizations of it. They are hybrid quantizations: rather than applying one precision to the whole network, each layer's precision is chosen from a measured importance score, so the bits go where they change the output most.

Model Overview

Attribute	Value
Base model	zai-org/GLM-5.2
Architecture	glm-dsa, Mixture-of-Experts
Parameters	743 B total, 41 B active per token
Layers	78
Hidden size	6,144
Attention heads	64
Experts	256 routed, 8 selected per token, shared expert
Context length	up to 1M tokens
Vocabulary	154,880
License	MIT

Attribute	Value
Base model	zai-org/GLM-5.2
Architecture	glm-dsa, Mixture-of-Experts
Parameters	743 B total, 41 B active per token
Layers	78
Hidden size	6,144
Attention heads	64
Experts	256 routed, 8 selected per token, shared expert
Context length	up to 1M tokens
Vocabulary	154,880
License	MIT

Quantizations

File	Type	Size	KLD vs bf16
GLM-5.2-Q5_K_M-*.gguf (4 parts)	Q5_K_M	578.95 GB	0.0214
GLM-5.2-Q4_K_M-*.gguf (4 parts)	Q4_K_M	492.81 GB	0.0353
GLM-5.2-IQ3_M-*.gguf (3 parts)	IQ3_M	356.29 GB	0.1731
GLM-5.2-IQ3_S-*.gguf (3 parts)	IQ3_S	352.27 GB	0.1843

File	Type	Size	KLD vs bf16
GLM-5.2-Q5_K_M-*.gguf (4 parts)	Q5_K_M	578.95 GB	0.0214
GLM-5.2-Q4_K_M-*.gguf (4 parts)	Q4_K_M	492.81 GB	0.0353
GLM-5.2-IQ3_M-*.gguf (3 parts)	IQ3_M	356.29 GB	0.1731
GLM-5.2-IQ3_S-*.gguf (3 parts)	IQ3_S	352.27 GB	0.1843

KL divergence measures how far a quantized model's next-token probabilities drift from those of the original bf16 model, averaged over every token of the evaluation text. The lower the value, the closer the quantized model is to the original; zero would mean identical output.

Evaluation

Every variant in this repository was evaluated against the original bf16 model for perplexity, KL divergence, top-token agreement, HellaSwag and WinoGrande, under identical conditions. The full results are maintained on the model's page in the Vultr Model Library:

https://docs.vultr.com/models/zai-glm-52/glm-5.2?tab=quantized

That page is the reference for these files. The summary above is an excerpt from it.

Run

llama-server -hf vultr/GLM-5.2-GGUF:Q4_K_M

The model's chat template is embedded in every file, so no extra prompt formatting is needed.

The repository also includes mtp-GLM-5.2-Q8_0.gguf, the model's multi-token prediction (MTP) head, for optional speculative decoding with any quant above:

llama-server -hf vultr/GLM-5.2-GGUF:Q4_K_M --spec-type draft-mtp

License

These files are derived from zai-org/GLM-5.2 and are distributed under the same license, MIT. The model is the work of Z.ai; Vultr produced the quantizations.