license: apache-2.0
base_model: AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16
language:
- en
- zh
- multilingual
library_name: transformers
pipeline_tag: text-generation
tags: - abliterated
- uncensored
- qwen3
- qwen3.6
- bf16
- bfloat16
- hybrid-attention
- mamba
- gated-deltanet
- abliterix
- fernflower-ssm-repair
- aeon
- a100
- h100
- pre-blackwell
- multi-gpu
- fine-tuning
- sm_80
- sm_90
- mlx
- mlx-my-repo
usermma/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16-mlx-4Bit
The Model mlx-community/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16-mlx-4Bit was converted to MLX format from AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16 using mlx-lm version 0.31.2.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16-mlx-4Bit")
prompt="hello"
if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)