Downloads · 30 days
360
12% of all-time downloads
mlx-community/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking-OptiQ-4bit
Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking-OptiQ-4bit is a text generation model from mlx-community. 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.
An OptiQ mixed-precision quantization of DavidAU/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking, built for Apple Silicon with MLX.
Downloads · 30 days
360
12% of all-time downloads
All-time downloads
3K
Public
Parameters
26.9B
18.8 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors18.7 GB · 100%
How the weights are stored.
U3226.9B · 100%
From the Hugging Face model README
An OptiQ mixed-precision quantization of DavidAU/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking, built for Apple Silicon with MLX.
OptiQ assigns a per-layer bit-width (4-bit or 8-bit here) instead of quantizing every layer the same, so the layers that need the precision keep it.
optiq_mixed_precision)qwen3_5 (Qwen3.6-27B):think / :no-think to the
model id, or set enable_thinking in the chat template, to control the
thinking channelThe per-layer allocation is reused from mlx-community/Qwen3.6-27B-OptiQ-4bit, which shares the same Qwen3.6-27B architecture, so no separate sensitivity pass was run for this fine-tune. This is a text-only quant of the language tower.
pip install mlx-optiq
optiq serve --model mlx-community/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking-OptiQ-4bit
Or load it directly with mlx_lm:
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking-OptiQ-4bit")
See mlx-optiq.com for the CLI, the local Lab UI, and the OptiQ Code agent. The base model and its behavior are DavidAU's; this repo only changes the quantization.