library_name: mlx
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3.6-27B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- heretic
- uncensored
- decensored
- abliterated
- mpoa
- mlx
base_model: llmfan46/Qwen3.6-27B-uncensored-heretic-v2
leonsarmiento/Qwen3.6-27B-uncensored-heretic-v2-3bit-mlx
This model leonsarmiento/Qwen3.6-27B-uncensored-heretic-v2-3bit-mlx was
converted to MLX format from llmfan46/Qwen3.6-27B-uncensored-heretic-v2
using mlx-lm version 0.31.2.
using mlx-lm version 0.31.2.
Quantization Details
The model uses mixed quantization:
- Embedding layers: 5-bit with group_size=64
- Prediction layers: 5-bit with group_size=64
- All other layers: 3-bit with group_size=64
This mixed precision approach provides a balance between compression and quality.
Use with mlx
pip install mlx-lm
Recommended Inference Parameters - Add to Jinja template on LM studio or Chat Template Kwargs on oMLX
Thinking Preserve ({%- set preserve_thinking = true %}):
- General tasks:
temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.05 - Coding tasks:
temperature=0.6, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=0.0, repetition_penalty=1.05
Instruct Mode ({%- set enable_thinking = false -%}):
- General tasks:
temperature=0.7, top_p=0.8, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.05 - Reasoning tasks:
temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.05
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("leonsarmiento/Qwen3.6-27B-uncensored-heretic-v2-3bit-mlx")
prompt = "hello"
if tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True, return_dict=False,
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)