What are the hardware requirements for GLM 4.7 Heretic?

GLM 4.7 Heretic requires approximately 224.9 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 4.7 Heretic and how large is it?

GLM 4.7 Heretic is developed by jarradh, a model with 352.8b 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 4.7 Heretic?

GLM 4.7 Heretic is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

GLM 4.7 Heretic by jarradh — VRAM 224.9GB, 198K context

Name: GLM 4.7 Heretic
Author: jarradh

GLM 4.7 Heretic is an open-source language model by jarradh. Features: 352.8b LLM, VRAM: 224.9GB, Context: 198K, License: mit, LLM Explorer Score: 0.22.

Base model:finetune:zai-org/gl... Base model:zai-org/glm-4.7 Conversational En Endpoints compatible Glm4 moe Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/jarradh/GLM-4.7-heretic

GLM 4.7 Heretic Parameters and Internals

LLM Name	GLM 4.7 Heretic
Repository 🤗	https://huggingface.co/jarradh/GLM-4.7-heretic
Base Model(s)	zai-org/GLM-4.7 zai-org/GLM-4.7
Model Size	352.8b
Required VRAM	224.9 GB
Updated	2026-08-06
Maintainer	jarradh
Model Type	glm4_moe
Model Files	5.0 GB: 1-of-142 5.0 GB: 2-of-142 5.0 GB: 3-of-142 5.0 GB: 4-of-142 5.0 GB: 5-of-142 5.0 GB: 6-of-142 5.0 GB: 7-of-142 5.0 GB: 8-of-142 5.0 GB: 9-of-142 5.0 GB: 10-of-142 5.0 GB: 11-of-142 5.0 GB: 12-of-142 5.0 GB: 13-of-142 5.0 GB: 14-of-142 5.0 GB: 15-of-142 5.0 GB: 16-of-142 5.0 GB: 17-of-142 5.0 GB: 18-of-142 5.0 GB: 19-of-142 5.0 GB: 20-of-142 5.0 GB: 21-of-142 5.0 GB: 22-of-142 5.0 GB: 23-of-142 5.0 GB: 24-of-142 5.0 GB: 25-of-142 5.0 GB: 26-of-142 5.0 GB: 27-of-142 5.0 GB: 28-of-142 5.0 GB: 29-of-142 5.0 GB: 30-of-142 5.0 GB: 31-of-142 5.0 GB: 32-of-142 5.0 GB: 33-of-142 5.0 GB: 34-of-142 5.0 GB: 35-of-142 5.0 GB: 36-of-142 4.9 GB: 37-of-142 5.0 GB: 38-of-142 5.0 GB: 39-of-142 5.0 GB: 40-of-142 5.0 GB: 41-of-142 5.0 GB: 42-of-142 5.0 GB: 43-of-142 5.0 GB: 44-of-142 5.0 GB: 45-of-142
Supported Languages	en zh
Model Architecture	Glm4MoeForCausalLM
License	mit
Context Length	202752
Model Max Length	202752
Transformers Version	4.57.3
Vocabulary Size	151552

Rank the GLM 4.7 Heretic 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 4.7 Heretic

Best Alternatives	Context / RAM	Downloads	Likes
GLM 4.7 Heretic	198K / 355.2 GB	461	1
GLM 4.6 Bf16	198K / 232.5 GB	20	3
GLM 4.7 6bit	198K / 239.2 GB	2060	1
GLM 4.7 4bit	198K / 197.4 GB	2822	5
GLM 4.7 8bit Gs32	198K / 198.6 GB	1937	4
GLM 4.6 4bit	198K / 197.4 GB	891	15
GLM 4.7 8bit	198K / 189.5 GB	144	3
GLM 4.6 8bit	198K / 189.5 GB	22	2
GLM 4.6 5bit	198K / 228.6 GB	125	3
GLM 4.6 Mlx 8bit Gs32	198K / 198.6 GB	12	2

Note: green Score (e.g. "73.2") means that the model is better than jarradh/GLM-4.7-heretic.