---
title: Nemotron-Cascade-2-30B-A3B-bf16
canonical_url: "https://www.modelscope.cn/models/mlx-community/Nemotron-Cascade-2-30B-A3B-bf16"
md_url: "https://www.modelscope.cn/models/mlx-community/Nemotron-Cascade-2-30B-A3B-bf16.md"
repository: mlx-community/Nemotron-Cascade-2-30B-A3B-bf16
last_updated: 2026-04-30
license: other
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - nemotron_h
architectures:
  - NemotronHForCausalLM
base_model:
  - nvidia/Nemotron-Cascade-2-30B-A3B
base_model_relation: finetune
parameters: 31.6B
tensor_type:
  - BF16
  - F32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 7
stars: 0
tags:
  - nvidia
  - nemotron-cascade-2
  - reasoning
  - general-purpose
  - SFT
  - RL
  - mlx
---

# Nemotron-Cascade-2-30B-A3B-bf16

> Nemotron-Cascade-2-30B-A3B-bf16 - mlx-community 在 ModelScope 开源的模型。mlx-community/Nemotron-Cascade-2-30B-A3B-bf16

mlx-community/Nemotron-Cascade-2-30B-A3B-bf16 是 ModelScope 魔搭社区上的 31.6B 参数text-generation模型，采用 other 许可，基于 nvidia/Nemotron-Cascade-2-30B-A3B 构建。

- **Repository**: mlx-community/Nemotron-Cascade-2-30B-A3B-bf16
- **License**: other
- **Tasks**: text-generation
- **Parameters**: 31.6B
- **Base model**: nvidia/Nemotron-Cascade-2-30B-A3B
- **Tags**: nvidia, nemotron-cascade-2, reasoning, general-purpose, SFT, RL, mlx
- **Downloads**: 7
- **Stars**: 0
- **Last updated**: 2026-04-30

Source: https://www.modelscope.cn/models/mlx-community/Nemotron-Cascade-2-30B-A3B-bf16

---

# mlx-community/Nemotron-Cascade-2-30B-A3B-bf16

This model [mlx-community/Nemotron-Cascade-2-30B-A3B-bf16](https://huggingface.co/mlx-community/Nemotron-Cascade-2-30B-A3B-bf16) was
converted to MLX format from [nvidia/Nemotron-Cascade-2-30B-A3B](https://huggingface.co/nvidia/Nemotron-Cascade-2-30B-A3B)
using mlx-lm version **0.31.3**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/Nemotron-Cascade-2-30B-A3B-bf16")

prompt = "hello"

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

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