Instructions to use assassindesign/gemma-4-e4b-8bit-fixed with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use assassindesign/gemma-4-e4b-8bit-fixed with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir gemma-4-e4b-8bit-fixed assassindesign/gemma-4-e4b-8bit-fixed
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,29 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
library_name: mlx
|
| 3 |
+
tags:
|
| 4 |
+
- mlx
|
| 5 |
+
- audio
|
| 6 |
+
- vision
|
| 7 |
+
- gemma-4
|
| 8 |
+
- quantized
|
| 9 |
---
|
| 10 |
+
|
| 11 |
+
# Gemma-4-E4B-it 8-bit MLX (Audio Supported / Fixed)
|
| 12 |
+
|
| 13 |
+
本模型是将官方原版 [google/gemma-4-e4b-it](https://huggingface.co/google/gemma-4-e4b-it) 转换为 Apple Silicon (Mac) 专用的 MLX 格式的 **8-bit 量化版本**。
|
| 14 |
+
|
| 15 |
+
该模型由 **mlx-vlm 版本 0.5.0** 转换生成。有关模型的架构、训练数据和预设能力的更多详细信息,请参阅[原始模型卡](https://huggingface.co/google/gemma-4-e4b-it)。
|
| 16 |
+
|
| 17 |
+
This model is an **8-bit quantized** MLX version of the original [google/gemma-4-e4b-it](https://huggingface.co/google/gemma-4-e4b-it), optimized for Apple Silicon Macs. It was converted using **mlx-vlm version 0.5.0**. For more details about the model architecture and training data, please refer to the original model card.
|
| 18 |
+
|
| 19 |
+
## 🌟 此版本的特别说明 (Important Fixes)
|
| 20 |
+
|
| 21 |
+
早期的 `mlx-vlm` 版本(如 0.4.x)在转换和量化 Gemma 4 时,存在破坏音频塔(Audio Tower)权重以及丢失 `feature_extractor` 的 Bug,导致模型在处理音频时出现“失聪”或无限输出网址乱码(如 `123456...`)的问题。
|
| 22 |
+
|
| 23 |
+
**本仓库的权重使用了最新的 mlx-vlm v0.5.0 重新转换,完美跳过了音频塔的损坏,保留了无损的音频特征提取能力!** 现在你可以直接在 Mac 上极速运行 Gemma 4 的原生语音识别(ASR)和音频对话功能。
|
| 24 |
+
|
| 25 |
+
## 💻 安装与依赖 (Requirements)
|
| 26 |
+
|
| 27 |
+
请确保你安装了最新版的 `mlx-vlm`(需要包含音频修复补丁):
|
| 28 |
+
```bash
|
| 29 |
+
pip install -U "mlx-vlm>=0.5.0" transformers
|