---
title: EXAONE-Deep-32B-mlx-8Bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/EXAONE-Deep-32B-mlx-8Bit"
md_url: "https://www.modelscope.cn/models/mlx-community/EXAONE-Deep-32B-mlx-8Bit.md"
repository: mlx-community/EXAONE-Deep-32B-mlx-8Bit
last_updated: 2025-03-21
license: other
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - exaone
architectures:
  - ExaoneForCausalLM
base_model:
  - LGAI-EXAONE/EXAONE-Deep-32B
base_model_relation: finetune
parameters: 9.0B
tensor_type:
  - F16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - ko
inference_backends:
  - "deploy_task text/emb"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 489
stars: 0
tags:
  - lg-ai
  - exaone
  - exaone-deep
  - mlx
  - mlx-my-repo
---

# EXAONE-Deep-32B-mlx-8Bit

> EXAONE-Deep-32B-mlx-8Bit - mlx-community 在 ModelScope 开源的模型。mlx-community/EXAONE-Deep-32B-mlx-8Bit

mlx-community/EXAONE-Deep-32B-mlx-8Bit 是 ModelScope 魔搭社区上的 9.0B 参数text-generation模型，采用 other 许可，基于 LGAI-EXAONE/EXAONE-Deep-32B 构建，可用 deploy_task text/emb、sglang 0.5.2、vllm 0.9.2 部署。

- **Repository**: mlx-community/EXAONE-Deep-32B-mlx-8Bit
- **License**: other
- **Tasks**: text-generation
- **Parameters**: 9.0B
- **Base model**: LGAI-EXAONE/EXAONE-Deep-32B
- **Inference backends**: deploy_task text/emb, sglang 0.5.2, vllm 0.9.2
- **Tags**: lg-ai, exaone, exaone-deep, mlx, mlx-my-repo
- **Downloads**: 489
- **Stars**: 0
- **Last updated**: 2025-03-21

Source: https://www.modelscope.cn/models/mlx-community/EXAONE-Deep-32B-mlx-8Bit

---

# mlx-community/EXAONE-Deep-32B-mlx-8Bit

The Model [mlx-community/EXAONE-Deep-32B-mlx-8Bit](https://huggingface.co/mlx-community/EXAONE-Deep-32B-mlx-8Bit) was converted to MLX format from [LGAI-EXAONE/EXAONE-Deep-32B](https://huggingface.co/LGAI-EXAONE/EXAONE-Deep-32B) using mlx-lm version **0.22.1**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/EXAONE-Deep-32B-mlx-8Bit")

prompt="hello"

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

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