What are the hardware requirements for GLM 4.5V?

GLM 4.5V requires approximately 212.7 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed GLM 4.5V and how large is it?

GLM 4.5V is developed by zai-org, a model with 107.7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does GLM 4.5V perform on standard benchmarks?

GLM 4.5V has the following published scores: LMSYS ELO 1354. Compare against reference models on this page or on the LLM Explorer leaderboards.

GLM 4.5V by zai-org — VRAM 212.7GB

Name: GLM 4.5V
Author: zai-org

GLM 4.5V is an open-source language model by zai-org. Features: 107.7b LLM, VRAM: 212.7GB, License: mit, LLM Explorer Score: 0.45, ELO: 1354.

Arxiv:2507.01006 Base model:finetune:zai-org/gl... Base model:zai-org/glm-4.5-air... Conversational En Endpoints compatible Glm4v moe Image-text-to-text Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/zai-org/GLM-4.5V

GLM 4.5V Benchmarks

LMSys ELO 1354 vs 1272 (so35) 6.4%

^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 4.5V Parameters and Internals

LLM Name	GLM 4.5V
Repository 🤗	https://huggingface.co/zai-org/GLM-4.5V
Base Model(s)	zai-org/GLM-4.5-Air-Base zai-org/GLM-4.5-Air-Base
Model Size	107.7b
Required VRAM	212.7 GB
Updated	2026-08-08
Maintainer	zai-org
Model Type	glm4v_moe
Model Files	5.2 GB: 1-of-46 4.7 GB: 2-of-46 4.7 GB: 3-of-46 4.7 GB: 4-of-46 4.7 GB: 5-of-46 4.7 GB: 6-of-46 4.7 GB: 7-of-46 4.7 GB: 8-of-46 4.7 GB: 9-of-46 4.7 GB: 10-of-46 4.7 GB: 11-of-46 4.7 GB: 12-of-46 4.7 GB: 13-of-46 4.7 GB: 14-of-46 4.7 GB: 15-of-46 4.7 GB: 16-of-46 4.7 GB: 17-of-46 4.7 GB: 18-of-46 4.7 GB: 19-of-46 4.7 GB: 20-of-46 4.7 GB: 21-of-46 4.7 GB: 22-of-46 4.7 GB: 23-of-46 4.7 GB: 24-of-46 4.7 GB: 25-of-46 4.7 GB: 26-of-46 4.7 GB: 27-of-46 4.7 GB: 28-of-46 4.7 GB: 29-of-46 4.7 GB: 30-of-46 4.7 GB: 31-of-46 4.7 GB: 32-of-46 4.7 GB: 33-of-46 4.7 GB: 34-of-46 4.7 GB: 35-of-46 4.7 GB: 36-of-46 4.7 GB: 37-of-46 4.7 GB: 38-of-46 4.7 GB: 39-of-46 4.7 GB: 40-of-46 4.7 GB: 41-of-46 4.7 GB: 42-of-46 4.7 GB: 43-of-46 4.7 GB: 44-of-46 5.4 GB: 45-of-46
Supported Languages	zh en
Model Architecture	Glm4vMoeForConditionalGeneration
License	mit
Transformers Version	4.57.1

Rank the GLM 4.5V 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

Quantized Models of the GLM 4.5V

Model	Likes	Downloads	VRAM
GLM 4.5V Abliterated 4bit	1	62	62 GB
GLM 4.5V Abliterated 8bit	1	56	115 GB

Best Alternatives to GLM 4.5V

Best Alternatives	Context / RAM	Downloads	Likes
GLM 4.6V	0K / 216.7 GB	9263	395
GLM 4.6V	0K / 216.7 GB	25	4