Downloads · 30 days
14
27% of all-time downloads
Decomposer-Qwen3-8B-MLX-8bit
afhverjuekki/Decomposer-Qwen3-8B-MLX-8bit
Decomposer-Qwen3-8B-MLX-8bit is a text generation model from afhverjuekki. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as apache-2.0.
8-bit MLX quantization of haiyewon/Decomposer-Qwen3-8B, a Qwen3-8B fine-tune that decompiles MIDI into Strudel pattern code. Runs locally on Apple Silicon via mlx-lm.
All-time downloads
52
Public
Parameters
8.2B
8.7 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors8.7 GB · 100%
Parameter types
How the weights are stored.
U328.2B · 100%
Try a prompt
Base models
Model card
From the Hugging Face model README
Decomposer-Qwen3-8B-MLX-8bit
8-bit MLX quantization of haiyewon/Decomposer-Qwen3-8B, a Qwen3-8B fine-tune that decompiles MIDI into Strudel pattern code. Runs locally on Apple Silicon via mlx-lm.
Converted with:
uvx --from mlx-lm mlx_lm.convert --hf-path haiyewon/Decomposer-Qwen3-8B \
--mlx-path Decomposer-Qwen3-8B-MLX-8bit -q --q-bits 8
Use with:
uvx --from mlx-lm mlx_lm.generate --model afhverjuekki/Decomposer-Qwen3-8B-MLX-8bit \
--prompt "..."
See the base model card for the Decomposer task format, training details, and evaluation.
- Task
- Text Generation
- Library
- mlx
- Type
- qwen3
- License
- apache-2.0
- Languages
- en
- Created
- Sep 1, 2026
- Updated
- Sep 1, 2026