license: other
license_name: swift-open-license-1.0
license_link: https://huggingface.co/ukisai/Swift-Qwen3.8-27b
base_model: ajgazin/Swift-Qwen3.8-27B-Uncensored-MTP
library_name: transformers
pipeline_tag: image-text-to-text
tags:
- abliterated
- uncensored
- qwen3_8
- mtp
- mlx
- mlx-my-repo
kylar/Swift-Qwen3.8-27B-Uncensored-MTP-mlx-8Bit
Created from: https://huggingface.co/spaces/mlx-community/mlx-my-repo
The Model kylar/Swift-Qwen3.8-27B-Uncensored-MTP-mlx-8Bit was converted to MLX format from ajgazin/Swift-Qwen3.8-27B-Uncensored-MTP using mlx-lm version 0.31.2.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("kylar/Swift-Qwen3.8-27B-Uncensored-MTP-mlx-8Bit")
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)