language: en
tags:
- obliteratus
- abliteration
- uncensored
- obliterate
- mlx
- mlx-my-repo
base_model: usermma/Atomight-V2.2-UltraThink-0.5B-abliterated
license: apache-2.0
usermma/Atomight-V2.2-UltraThink-0.5B-abliterated-mlx-6Bit
The Model usermma/Atomight-V2.2-UltraThink-0.5B-abliterated-mlx-6Bit was converted to MLX format from usermma/Atomight-V2.2-UltraThink-0.5B-abliterated using mlx-lm version 0.31.2.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("usermma/Atomight-V2.2-UltraThink-0.5B-abliterated-mlx-6Bit")
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)