---
title: VibeVoice-Realtime-0.5B-4bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/VibeVoice-Realtime-0.5B-4bit"
md_url: "https://www.modelscope.cn/models/mlx-community/VibeVoice-Realtime-0.5B-4bit.md"
repository: mlx-community/VibeVoice-Realtime-0.5B-4bit
last_updated: 2025-12-16
license: mit
pipeline_tag: text-to-speech
tasks:
  - text-to-speech
model_type:
  - vibevoice_streaming
architectures:
  - VibeVoiceStreamingForConditionalGenerationInference
base_model:
  - Qwen/Qwen2.5-0.5B
base_model_relation: quantized
parameters: 194.3M
tensor_type:
  - F16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 159
stars: 0
tags:
  - "Realtime TTS"
  - "Streaming text input"
  - "Long-form speech generation"
  - mlx
---

# VibeVoice-Realtime-0.5B-4bit

> VibeVoice-Realtime-0.5B-4bit - mlx-community 在 ModelScope 开源的模型。mlx-community/VibeVoice-Realtime-0.5B-4bit This model was converted to MLX format from microsoft/VibeVoice-Realtime-0.5B using mlx-audio version 0.2.6. Refer to the original model card for more…

mlx-community/VibeVoice-Realtime-0.5B-4bit 是 ModelScope 魔搭社区上的 194.3M 参数text-to-speech模型，采用 mit 许可，基于 Qwen/Qwen2.5-0.5B 构建。

- **Repository**: mlx-community/VibeVoice-Realtime-0.5B-4bit
- **License**: mit
- **Tasks**: text-to-speech
- **Parameters**: 194.3M
- **Base model**: Qwen/Qwen2.5-0.5B
- **Tags**: Realtime TTS, Streaming text input, Long-form speech generation, mlx
- **Downloads**: 159
- **Stars**: 0
- **Last updated**: 2025-12-16

Source: https://www.modelscope.cn/models/mlx-community/VibeVoice-Realtime-0.5B-4bit

---

# mlx-community/VibeVoice-Realtime-0.5B-4bit
This model was converted to MLX format from [`microsoft/VibeVoice-Realtime-0.5B`](https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B) using mlx-audio version **0.2.6**.
Refer to the [original model card](https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B) for more details on the model.
## Use with mlx

```bash
pip install -U mlx-audio
```

```bash
python -m mlx_audio.tts.generate --model mlx-community/VibeVoice-Realtime-0.5B-4bit --text "Hello, this is VibeVoice real-time 0.5B model." --voice en-Emma_woman
```