What are the hardware requirements for GLM 5.2 Mxfp4?

GLM 5.2 Mxfp4 requires approximately 233.5 GB of VRAM and supports a context window of 1024K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed GLM 5.2 Mxfp4 and how large is it?

GLM 5.2 Mxfp4 is developed by mlx-community, a model with 743.4b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does GLM 5.2 Mxfp4 perform on standard benchmarks?

GLM 5.2 Mxfp4 has the following published scores: LMSYS ELO 1471. Compare against reference models on this page or on the LLM Explorer leaderboards.

GLM 5.2 Mxfp4 by mlx-community — VRAM 233.5GB, 1024K context

Name: GLM 5.2 Mxfp4
Author: mlx-community

GLM 5.2 Mxfp4 is an open-source language model by mlx-community. Features: 743.4b LLM, VRAM: 233.5GB, Context: 1024K, MoE, LLM Explorer Score: 0.47, ELO: 1471.

4-bit Base model:quantized:zai-org/g... Base model:zai-org/glm-5.2-fp8 Conversational En Glm moe dsa Mlx Moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mlx-community/GLM-5.2-mxfp4

GLM 5.2 Mxfp4 Benchmarks

LMSys ELO 1471 vs 1272 (so35) 15.6%

^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 5.2 Mxfp4 Parameters and Internals

LLM Name	GLM 5.2 Mxfp4
Repository 🤗	https://huggingface.co/mlx-community/GLM-5.2-mxfp4
Base Model(s)	zai-org/GLM-5.2-FP8 zai-org/GLM-5.2-FP8
Model Size	743.4b
Required VRAM	233.5 GB
Updated	2026-08-02
Maintainer	mlx-community
Model Type	glm_moe_dsa
Model Files	4.7 GB: 1-of-76 5.2 GB: 2-of-76 5.2 GB: 3-of-76 5.2 GB: 4-of-76 5.2 GB: 5-of-76 5.2 GB: 6-of-76 5.2 GB: 7-of-76 5.2 GB: 8-of-76 5.2 GB: 9-of-76 5.2 GB: 10-of-76 5.2 GB: 11-of-76 5.2 GB: 12-of-76 5.2 GB: 13-of-76 5.2 GB: 14-of-76 5.2 GB: 15-of-76 5.2 GB: 16-of-76 5.2 GB: 17-of-76 5.2 GB: 18-of-76 5.2 GB: 19-of-76 5.2 GB: 20-of-76 5.2 GB: 21-of-76 5.2 GB: 22-of-76 5.2 GB: 23-of-76 5.2 GB: 24-of-76 5.2 GB: 25-of-76 5.2 GB: 26-of-76 5.2 GB: 27-of-76 5.2 GB: 28-of-76 5.2 GB: 29-of-76 5.2 GB: 30-of-76 5.2 GB: 31-of-76 5.2 GB: 32-of-76 5.2 GB: 33-of-76 5.2 GB: 34-of-76 5.2 GB: 35-of-76 5.2 GB: 36-of-76 5.2 GB: 37-of-76 5.2 GB: 38-of-76 5.2 GB: 39-of-76 5.2 GB: 40-of-76 5.2 GB: 41-of-76 5.2 GB: 42-of-76 5.2 GB: 43-of-76 5.2 GB: 44-of-76 5.2 GB: 45-of-76
Supported Languages	en
Model Architecture	GlmMoeDsaForCausalLM
Context Length	1048576
Model Max Length	1048576
Transformers Version	5.12.0
Vocabulary Size	154880

Rank the GLM 5.2 Mxfp4 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 5.2 Mxfp4

Best Alternatives	Context / RAM	Downloads	Likes
GLM 5.2 MLX Nvfp4	1024K / 207.5 GB	4741	3
Aesop V1	1024K / 210.9 GB	905	4
GLM 5.2 DQ4plus Q8	1024K / 184.6 GB	9208	16
GLM 5.2 4bit	1024K / 207.5 GB	9403	5