Downloads · 30 days
23
12% of all-time downloads
mlx-qwen3-asr-1.7b-8bit
Six666/mlx-qwen3-asr-1.7b-8bit
mlx-qwen3-asr-1.7b-8bit is a machine learning model from Six666. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for mlx-qwen3-asr. The card lists the license as apache-2.0.
8-bit MLX-quantized conversion of Qwen/Qwen3-ASR-1.7B, for use with the mlx-qwen3-asr package on Apple Silicon.
All-time downloads
191
Public
Repo size
2.5 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors2.5 GB · 100%
At a glance
- Library
- mlx-qwen3-asr
- License
- apache-2.0
- Model type
- qwen3_asr
- Access
- Public
- Created
- Jun 19, 2026
- Updated
- Jun 19, 2026
- SHA
- cecaf5b0
Base models
Model card
From the Hugging Face model README
Six666/mlx-qwen3-asr-1.7b-8bit
8-bit MLX-quantized conversion of Qwen/Qwen3-ASR-1.7B,
for use with the mlx-qwen3-asr package on Apple Silicon.
- bits: 8
- group_size: 64
- dtype (pre-quant): float16
- quantization: MLX affine group quantization (
nn.quantize)
Usage
from mlx_qwen3_asr import transcribe
result = transcribe("audio.wav", model="Six666/mlx-qwen3-asr-1.7b-8bit")
print(result.text)
mlx-qwen3-asr audio.wav --model Six666/mlx-qwen3-asr-1.7b-8bit
Note: produced with
mlx-qwen3-asr's own converter. Themlx-community/Qwen3-ASR-*quantized repos use mlx-audio's layer naming and will not load with this package.
- Library
- mlx-qwen3-asr
- Type
- qwen3_asr
- License
- apache-2.0
- Created
- Jun 19, 2026
- Updated
- Jun 19, 2026