What are the hardware requirements for GLM 5 4bit?

GLM 5 4bit requires approximately 210.3 GB of VRAM and supports a context window of 198K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed GLM 5 4bit and how large is it?

GLM 5 4bit is developed by mlx-community, a model with 743.9b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does GLM 5 4bit perform on standard benchmarks?

GLM 5 4bit has the following published scores: LMSYS ELO 1456. Compare against reference models on this page or on the LLM Explorer leaderboards.

GLM 5 4bit by mlx-community — VRAM 210.3GB, 198K context

Name: GLM 5 4bit
Author: mlx-community

GLM 5 4bit is an open-source language model by mlx-community. Features: 743.9b LLM, VRAM: 210.3GB, Context: 198K, License: mit, MoE, Quantized, LLM Explorer Score: 0.39, ELO: 1456.

4-bit 4bit Base model:quantized:zai-org/g... Base model:zai-org/glm-5 Conversational En Glm moe dsa Mlx Moe Quantized Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/mlx-community/GLM-5-4bit

GLM 5 4bit Benchmarks

LMSys ELO 1456 vs 1272 (so35) 14.5%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

GLM 5 4bit Parameters and Internals

LLM Name	GLM 5 4bit
Repository 🤗	https://huggingface.co/mlx-community/GLM-5-4bit
Base Model(s)	GLM 5 zai-org/GLM-5
Model Size	743.9b
Required VRAM	210.3 GB
Updated	2026-07-30
Maintainer	mlx-community
Model Type	glm_moe_dsa
Model Files	4.9 GB: 1-of-91 5.4 GB: 2-of-91 4.0 GB: 3-of-91 5.4 GB: 4-of-91 4.0 GB: 5-of-91 5.3 GB: 6-of-91 3.9 GB: 7-of-91 5.4 GB: 8-of-91 4.0 GB: 9-of-91 5.4 GB: 10-of-91 4.0 GB: 11-of-91 5.3 GB: 12-of-91 3.9 GB: 13-of-91 5.4 GB: 14-of-91 4.0 GB: 15-of-91 5.4 GB: 16-of-91 4.0 GB: 17-of-91 5.3 GB: 18-of-91 3.9 GB: 19-of-91 5.4 GB: 20-of-91 4.0 GB: 21-of-91 5.4 GB: 22-of-91 4.0 GB: 23-of-91 5.3 GB: 24-of-91 3.9 GB: 25-of-91 5.4 GB: 26-of-91 4.0 GB: 27-of-91 5.4 GB: 28-of-91 4.0 GB: 29-of-91 5.3 GB: 30-of-91 3.9 GB: 31-of-91 5.4 GB: 32-of-91 4.0 GB: 33-of-91 5.4 GB: 34-of-91 4.0 GB: 35-of-91 5.3 GB: 36-of-91 3.9 GB: 37-of-91 5.4 GB: 38-of-91 4.0 GB: 39-of-91 5.4 GB: 40-of-91 4.0 GB: 41-of-91 5.3 GB: 42-of-91 3.9 GB: 43-of-91 5.4 GB: 44-of-91 4.0 GB: 45-of-91
Supported Languages	en zh
Quantization Type	4bit
Model Architecture	GlmMoeDsaForCausalLM
License	mit
Context Length	202752
Model Max Length	202752
Transformers Version	5.0.2.dev0
Vocabulary Size	154880

Rank the GLM 5 4bit capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to GLM 5 4bit

Best Alternatives	Context / RAM	Downloads	Likes
GLM 5.1 DQ4plus Q8	198K / 186.2 GB	1692	6
GLM 5 MXFP4 Q8	198K / 241.8 GB	172	1
GLM 5.1 MXFP4 Q8	198K / 241.8 GB	2733	4
GLM 5.1	198K / 220.3 GB	3662	5
GLM 5	198K / 220.3 GB	272	2

Note: green Score (e.g. "73.2") means that the model is better than mlx-community/GLM-5-4bit.