What are the hardware requirements for GLM 5.3 AWQ INT4?

GLM 5.3 AWQ INT4 requires approximately 214.9 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 5.3 AWQ INT4 and how large is it?

GLM 5.3 AWQ INT4 is developed by cyankiwi, a model with 151.7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does GLM 5.3 AWQ INT4 perform on standard benchmarks?

GLM 5.3 AWQ INT4 has the following published scores: LMSYS ELO 1484. Compare against reference models on this page or on the LLM Explorer leaderboards.

GLM 5.3 AWQ INT4 by cyankiwi — VRAM 214.9GB, 1024K context

Name: GLM 5.3 AWQ INT4
Author: cyankiwi

GLM 5.3 AWQ INT4 is an open-source language model by cyankiwi. Features: 151.7b LLM, VRAM: 214.9GB, Context: 1024K, License: other, MoE, Quantized, LLM Explorer Score: 0.41, ELO: 1484.

Arxiv:2602.15763 Awq Base model:quantized:zai-org/g... Base model:zai-org/glm-5.3 Compressed-tensors Conversational En Endpoints compatible Glm moe dsa Moe Quantized Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/cyankiwi/GLM-5.3-AWQ-INT4

GLM 5.3 AWQ INT4 Benchmarks

LMSys ELO 1484 vs 1272 (so35) 16.7%

^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 5.3 AWQ INT4 Parameters and Internals

LLM Name	GLM 5.3 AWQ INT4
Repository 🤗	https://huggingface.co/cyankiwi/GLM-5.3-AWQ-INT4
Base Model(s)	zai-org/GLM-5.3 zai-org/GLM-5.3
Model Size	151.7b
Required VRAM	214.9 GB
Updated	2026-09-08
Maintainer	cyankiwi
Model Type	glm_moe_dsa
Model Files	5.0 GB: 1-of-98 5.0 GB: 2-of-98 5.0 GB: 3-of-98 5.0 GB: 4-of-98 4.9 GB: 5-of-98 5.0 GB: 6-of-98 5.0 GB: 7-of-98 5.0 GB: 8-of-98 5.0 GB: 9-of-98 5.0 GB: 10-of-98 5.0 GB: 11-of-98 5.0 GB: 12-of-98 5.0 GB: 13-of-98 5.0 GB: 14-of-98 5.0 GB: 15-of-98 5.0 GB: 16-of-98 5.0 GB: 17-of-98 5.0 GB: 18-of-98 5.0 GB: 19-of-98 5.0 GB: 20-of-98 5.0 GB: 21-of-98 5.0 GB: 22-of-98 5.0 GB: 23-of-98 5.0 GB: 24-of-98 5.0 GB: 25-of-98 5.0 GB: 26-of-98 5.0 GB: 27-of-98 5.0 GB: 28-of-98 5.0 GB: 29-of-98 5.0 GB: 30-of-98 5.0 GB: 31-of-98 5.0 GB: 32-of-98 5.0 GB: 33-of-98 5.0 GB: 34-of-98 5.0 GB: 35-of-98 5.0 GB: 36-of-98 5.0 GB: 37-of-98 5.0 GB: 38-of-98 5.0 GB: 39-of-98 5.0 GB: 40-of-98 5.0 GB: 41-of-98 5.0 GB: 42-of-98 5.0 GB: 43-of-98
Supported Languages	en zh
AWQ Quantization	Yes
Quantization Type	awq
Model Architecture	GlmMoeDsaForCausalLM
License	other
Context Length	1048576
Model Max Length	1048576
Transformers Version	5.16.0.dev0
Vocabulary Size	154880

Rank the GLM 5.3 AWQ INT4 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 5.3 AWQ INT4

Model	Likes	Downloads	VRAM
Altar 1	215	2350	1 GB
Altar 1	1	112	1 GB