GLM-4-9B-0414-GGUF

GLM-4-9B-0414-GGUF

Generates text in Chinese and English from GGUF files of the 9.4B GLM-4-9B-0414 model, for people who want to run a GLM model on their own computer.

Made by unsloth on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: This repository holds GGUF files of GLM-4-9B-0414, a 9.4B text generation model from the GLM family, packaged by unsloth. GGUF is the file format that local runners such as llama.cpp read. The card is mostly the family overview, which covers the larger 32B models and reasoning variants.

Who it suits: It suits people who want a mid-size Chinese and English model on their own hardware, and developers who test it for chat, code or tool calling.

Industry | Who uses it | How they use it | Example
Software | Developers | Run a Chinese and English model locally and try it on code tasks | A developer asks it to write a small Python function and reviews the result
Business | Teams working in two languages | Draft and revise text in Chinese or English | A team asks for an English summary of a Chinese memo
Research | Model testers | Compare this 9B model with others of similar size | A tester runs one prompt set against several local models

If you want a hosted service with no downloads, you can skip it.

Where to use it:

• llama.cpp and other local runners that read GGUF files
• Your own Windows, macOS or Linux computer, after downloading the files with Python

Use case | Who | What they get back
Local chat | Individual users | Text replies in Chinese or English
Code help | Developer | Draft code and explanations for a task
Text drafting | Business team | A first draft or a rewrite of a document

How to set it up:

• Python on your computer, to run pip and the Hugging Face tools
• Disk space, about 4.7 GB for a 4-bit copy by Hugging Face's estimate
• A runner that loads GGUF files, such as llama.cpp

1. Install the Hugging Face hub package.
2. Download the repository files.
3. Load a GGUF file in your runner. With llama.cpp, add the jinja flag so the system prompt works.

pip install -U huggingface_hub
hf download unsloth/GLM-4-9B-0414-GGUF

How to use it: You prompt the model in Chinese or English and it writes text back. The one usage note on the card is for llama.cpp, where the jinja flag switches on the system prompt. The only command shown in full is the download step.

hf download unsloth/GLM-4-9B-0414-GGUF

Tool | What it does
hf download | Fetches the repository files to your computer
llama.cpp with --jinja | Runs the GGUF file and enables the system prompt

Example: You ask, "Write a Python function that checks whether a string is a palindrome." The model replies with the function and usually a short explanation of how it works.

For newcomers and experts: GGUF is a file format for running a model on your own computer, and this model writes text from the prompt you give it. A small first step is to download the files, load one in llama.cpp with the jinja flag and ask a short question.

For experts, the base is GLM-4-9B-0414 with 9.4B parameters. The card describes the GLM 0414 family as trained with human preference alignment, rejection sampling and reinforcement learning for instruction following, code and function calling. It gives those details for the 32B model, so check how far they carry to this size. The page says the unsloth Dynamic 2.0 method gives better accuracy than other quants.

THUDM releases GLM-4-9B-0414
unsloth converts it to GGUF files
You download the files with hf
llama.cpp or another runner loads one
You prompt it and it writes text

Good to know:

• Languages: Chinese and English
• Size: 9.4B parameters, about 4.7 GB at 4-bit by Hugging Face's estimate
• Card scope: most of the text describes the 32B and reasoning models in the family, not this 9B one
• License: MIT
• Cost to run: none named, apart from your own hardware
• Data it reaches: downloading uses the web and writes files to your disk, and the source names no other access
• Not for: llama.cpp runs without the jinja flag if you need the system prompt, since the card says the flag is required for it

On Hugging Face, 13 November 2026

• Task: text generation
• Library: transformers
• Parameters: 9.4B
• Weights: about 4.7 GB at 4-bit (estimate)
• Formats: GGUF
• Downloads in the last 30 days: 13,323
• Likes: 33
• License: mit
• Languages: zh, en
• Based on: THUDM/GLM-4-9B-0414
• Tags: transformers, gguf, glm4, text-generation, unsloth, conversational

Links

• Model: https://huggingface.co/unsloth/GLM-4-9B-0414-GGUF
• Files: https://huggingface.co/unsloth/GLM-4-9B-0414-GGUF/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download unsloth/GLM-4-9B-0414-GGUF

Then follow the usage in its model card. Nothing in this kit runs by itself.