What are the hardware requirements for GLM 5 MXFP4 Q8?

GLM 5 MXFP4 Q8 requires approximately 241.8 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 MXFP4 Q8 and how large is it?

GLM 5 MXFP4 Q8 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 MXFP4 Q8 perform on standard benchmarks?

GLM 5 MXFP4 Q8 has the following published scores: LMSYS ELO 1457. Compare against reference models on this page or on the LLM Explorer leaderboards.

GLM 5 MXFP4 Q8 by mlx-community — VRAM 241.8GB, 198K context

Name: GLM 5 MXFP4 Q8
Author: mlx-community

GLM 5 MXFP4 Q8 is an open-source language model by mlx-community. Features: 743.9b LLM, VRAM: 241.8GB, Context: 198K, License: mit, MoE, Quantized, LLM Explorer Score: 0.38, ELO: 1457.

4-bit Base model:quantized:zai-org/g... Base model:zai-org/glm-5 Conversational En Glm moe dsa Mlx Moe Mxfp4 Q8 Quantized Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/mlx-community/GLM-5-MXFP4-Q8

GLM 5 MXFP4 Q8 Benchmarks

LMSys ELO 1457 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 MXFP4 Q8 Parameters and Internals

LLM Name	GLM 5 MXFP4 Q8
Repository 🤗	https://huggingface.co/mlx-community/GLM-5-MXFP4-Q8
Base Model(s)	GLM 5 zai-org/GLM-5
Model Size	743.9b
Required VRAM	241.8 GB
Updated	2026-08-10
Maintainer	mlx-community
Model Type	glm_moe_dsa
Model Files	4.2 GB: 1-of-76 5.4 GB: 2-of-76 5.4 GB: 3-of-76 5.4 GB: 4-of-76 5.4 GB: 5-of-76 5.4 GB: 6-of-76 5.4 GB: 7-of-76 5.4 GB: 8-of-76 5.4 GB: 9-of-76 5.4 GB: 10-of-76 5.4 GB: 11-of-76 5.4 GB: 12-of-76 5.4 GB: 13-of-76 5.4 GB: 14-of-76 5.4 GB: 15-of-76 5.4 GB: 16-of-76 5.4 GB: 17-of-76 5.4 GB: 18-of-76 5.4 GB: 19-of-76 5.4 GB: 20-of-76 5.4 GB: 21-of-76 5.4 GB: 22-of-76 5.4 GB: 23-of-76 5.4 GB: 24-of-76 5.4 GB: 25-of-76 5.4 GB: 26-of-76 5.4 GB: 27-of-76 5.4 GB: 28-of-76 5.4 GB: 29-of-76 5.4 GB: 30-of-76 5.4 GB: 31-of-76 5.4 GB: 32-of-76 5.4 GB: 33-of-76 5.4 GB: 34-of-76 5.4 GB: 35-of-76 5.4 GB: 36-of-76 5.4 GB: 37-of-76 5.4 GB: 38-of-76 5.4 GB: 39-of-76 5.4 GB: 40-of-76 5.4 GB: 41-of-76 5.4 GB: 42-of-76 5.4 GB: 43-of-76 5.4 GB: 44-of-76 5.4 GB: 45-of-76
Supported Languages	en zh
Quantization Type	q8
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 MXFP4 Q8 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 MXFP4 Q8

Best Alternatives	Context / RAM	Downloads	Likes
GLM 5.1 DQ4plus Q8	198K / 186.2 GB	1692	6
GLM 5 4bit	198K / 210.3 GB	1115	7
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-MXFP4-Q8.