Downloads · 30 days
9
4% of all-time downloads
Qwopus3.6-27B-v2-MLX-4bit
zaydiscold/Qwopus3.6-27B-v2-MLX-4bit
Qwopus3.6-27B-v2-MLX-4bit is a machine learning model from zaydiscold. 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. The card lists the license as apache-2.0.
4-bit MLX quantization of Jackrong/Qwopus3.6-27B-v2. Group size 64. The standard daily-driver quant for Apple Silicon.
All-time downloads
249
Public
Parameters
27.8B
15.8 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors15.8 GB · 100%
Parameter types
How the weights are stored.
U3227.6B · 100%
Base models
Model card
From the Hugging Face model README
Qwopus3.6-27B-v2 MLX 4bit
4-bit MLX quantization of Jackrong/Qwopus3.6-27B-v2. Group size 64. The standard daily-driver quant for Apple Silicon.
Use
pip install mlx-lm
mlx_lm.generate --model zaydiscold/Qwopus3.6-27B-v2-MLX-4bit \
--prompt "Explain quantum entanglement" --max-tokens 200 --temp 0.8
Author's temperature recommendation: 0.75–1.0.
The full MLX ladder
| Variant | Repo |
|---|---|
| MLX bf16 | Qwopus3.6-27B-v2-MLX-bf16 |
| MLX 4bit (this repo) | this |
More variants (8bit, 6bit, 3bit, group-size sweeps) shipping shortly.
Credits
- Source:
Jackrong/Qwopus3.6-27B-v2by Jackrong - MLX 4-bit conversion by zaydiscold
Mix brought by the NOTORIOUS MLX.
- Library
- mlx
- Type
- qwen3_5
- License
- apache-2.0
- Languages
- en, zh, ja, ko, es, ru
- Created
- May 22, 2026
- Updated
- May 22, 2026