---
title: EXAONE-4.0-32B-bf16
canonical_url: "https://www.modelscope.cn/models/mlx-community/EXAONE-4.0-32B-bf16"
md_url: "https://www.modelscope.cn/models/mlx-community/EXAONE-4.0-32B-bf16.md"
repository: mlx-community/EXAONE-4.0-32B-bf16
last_updated: 2025-07-18
license: other
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - exaone4
architectures:
  - Exaone4ForCausalLM
base_model:
  - LGAI-EXAONE/EXAONE-4.0-32B
base_model_relation: finetune
parameters: 32.0B
tensor_type:
  - BF16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - ko
  - es
downloads: 148
stars: 0
tags:
  - lg-ai
  - exaone
  - exaone-4.0
  - mlx
---

# EXAONE-4.0-32B-bf16

> EXAONE-4.0-32B-bf16 - mlx-community 在 ModelScope 开源的模型。mlx-community/EXAONE-4.0-32B-4bit

mlx-community/EXAONE-4.0-32B-bf16 是 ModelScope 魔搭社区上的 32.0B 参数text-generation模型，采用 other 许可，基于 LGAI-EXAONE/EXAONE-4.0-32B 构建。

- **Repository**: mlx-community/EXAONE-4.0-32B-bf16
- **License**: other
- **Tasks**: text-generation
- **Parameters**: 32.0B
- **Base model**: LGAI-EXAONE/EXAONE-4.0-32B
- **Tags**: lg-ai, exaone, exaone-4.0, mlx
- **Downloads**: 148
- **Stars**: 0
- **Last updated**: 2025-07-18

Source: https://www.modelscope.cn/models/mlx-community/EXAONE-4.0-32B-bf16

---

# mlx-community/EXAONE-4.0-32B-4bit

This model [mlx-community/EXAONE-4.0-32B-4bit](https://huggingface.co/mlx-community/EXAONE-4.0-32B-4bit) was
converted to MLX format from [LGAI-EXAONE/EXAONE-4.0-32B](https://huggingface.co/LGAI-EXAONE/EXAONE-4.0-32B)
using mlx-lm version **0.26.0**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/EXAONE-4.0-32B-4bit")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```