What are the hardware requirements for GLM 5.3 FP8?

GLM 5.3 FP8 requires approximately 120.7 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 FP8 and how large is it?

GLM 5.3 FP8 is developed by JANGQ-AI, a model with 753.3b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate GLM 5.3 FP8?

GLM 5.3 FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

GLM 5.3 FP8 by JANGQ-AI — VRAM 120.7GB, 1024K context

Name: GLM 5.3 FP8
Author: JANGQ-AI

GLM 5.3 FP8 is an open-source language model by JANGQ-AI. Features: 753.3b LLM, VRAM: 120.7GB, Context: 1024K, License: other, MoE, LLM Explorer Score: 0.28.

Base model:quantized:zai-org/g... Base model:zai-org/glm-5.3 Conversational En Endpoints compatible Fp8 Glm Glm-moe-dsa Glm moe dsa H200 Hopper Jang Moe Mtp Region:us Safetensors Sharded Tensorflow Vllm Zh

Model Card on HF 🤗: https://huggingface.co/JANGQ-AI/GLM-5.3-FP8

GLM 5.3 FP8 Parameters and Internals

LLM Name	GLM 5.3 FP8
Repository 🤗	https://huggingface.co/JANGQ-AI/GLM-5.3-FP8
Base Model(s)	zai-org/GLM-5.3 zai-org/GLM-5.3
Model Size	753.3b
Required VRAM	120.7 GB
Updated	2026-08-31
Maintainer	JANGQ-AI
Model Type	glm_moe_dsa
Model Files	4.6 GB: 1-of-282 2.7 GB: 2-of-282 2.7 GB: 3-of-282 2.7 GB: 4-of-282 2.7 GB: 5-of-282 2.7 GB: 6-of-282 2.7 GB: 7-of-282 2.7 GB: 8-of-282 2.7 GB: 9-of-282 2.7 GB: 10-of-282 2.7 GB: 11-of-282 2.7 GB: 12-of-282 2.7 GB: 13-of-282 2.7 GB: 14-of-282 2.7 GB: 15-of-282 2.7 GB: 16-of-282 2.7 GB: 17-of-282 2.7 GB: 18-of-282 2.7 GB: 19-of-282 2.7 GB: 20-of-282 2.7 GB: 21-of-282 2.7 GB: 22-of-282 2.7 GB: 23-of-282 2.7 GB: 24-of-282 2.7 GB: 25-of-282 2.7 GB: 26-of-282 2.7 GB: 27-of-282 2.7 GB: 28-of-282 2.7 GB: 29-of-282 2.7 GB: 30-of-282 2.7 GB: 31-of-282 2.7 GB: 32-of-282 2.7 GB: 33-of-282 2.7 GB: 34-of-282 2.7 GB: 35-of-282 2.7 GB: 36-of-282 2.7 GB: 37-of-282 2.7 GB: 38-of-282 2.7 GB: 39-of-282 2.7 GB: 40-of-282 2.7 GB: 41-of-282 2.7 GB: 42-of-282 2.7 GB: 43-of-282 2.7 GB: 44-of-282
Supported Languages	en zh
Model Architecture	GlmMoeDsaForCausalLM
License	other
Context Length	1048576
Model Max Length	1048576
Transformers Version	5.15.0
Vocabulary Size	154880

Rank the GLM 5.3 FP8 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 5.3 FP8

Best Alternatives	Context / RAM	Downloads	Likes
GLM 5.3	1024K / 232.2 GB	8804	1255
GLM 5.2	1024K / 232 GB	2391730	4875
GLM 5.3 BF16	1024K / 237.4 GB	482	24
GLM 5.3	1024K / 237.4 GB	379	6
GLM 5.3 W4A16	1024K / 73.9 GB	268	1
GLM 5.3	1024K / 232.2 GB	118	1
GLM 5.2 MXFP4xMXFP8	1024K / 68.1 GB	296	2
GLM 5.2	1024K / 237.4 GB	1671	26
MISS V0	1024K / 232 GB	205	1
GLM 5.2	1024K / 232 GB	116	1

Note: green Score (e.g. "73.2") means that the model is better than JANGQ-AI/GLM-5.3-FP8.