---
title: GLM-5.3-4bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/GLM-5.3-4bit"
md_url: "https://www.modelscope.cn/models/mlx-community/GLM-5.3-4bit.md"
repository: mlx-community/GLM-5.3-4bit
last_updated: 2026-09-03
license: other
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - glm_moe_dsa
architectures:
  - GlmMoeDsaForCausalLM
base_model:
  - zai-org/GLM-5.3-BF16
base_model_relation: quantized
parameters: 116.3B
tensor_type:
  - BF16
  - F32
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - zh
downloads: 50
stars: 0
tags:
  - mlx
  - transformers
---

# GLM-5.3-4bit

> GLM-5.3-4bit - mlx-community 在 ModelScope 开源的模型。mlx-community/GLM-5.3-4bit

mlx-community/GLM-5.3-4bit 是 ModelScope 魔搭社区上的 116.3B 参数text-generation模型，采用 other 许可，基于 zai-org/GLM-5.3-BF16 构建。

- **Repository**: mlx-community/GLM-5.3-4bit
- **License**: other
- **Tasks**: text-generation
- **Parameters**: 116.3B
- **Base model**: zai-org/GLM-5.3-BF16
- **Tags**: mlx, transformers
- **Downloads**: 50
- **Stars**: 0
- **Last updated**: 2026-09-03

Source: https://www.modelscope.cn/models/mlx-community/GLM-5.3-4bit

---

# mlx-community/GLM-5.3-4bit

This model [mlx-community/GLM-5.3-4bit](https://huggingface.co/mlx-community/GLM-5.3-4bit) was converted to MLX format from [zai-org/GLM-5.3-BF16](https://huggingface.co/zai-org/GLM-5.3-BF16) using mlx-lm version **0.31.3** (with PR #1410).

Note that this quant is using the GLM-5.3-BF16 as base. Testing various quant recipes, these often start to overthink and redoing "decisions". The standard 4-bit quant is stable and fast.

This is created for people using a single Apple Mac Studio M3 Ultra with 512 GB. The 4-bit version of GLM-5.3 fits comfortably.

You can find more similar MLX model quants for Apple Mac Studio with 512 GB at https://huggingface.co/bibproj

```bash
pip install mlx-lm

mlx_lm.generate --model mlx-community/GLM-5.3-4bit --prompt "Hi"
```

Enjoy!