Views
No views yet
Xiaolong is a small, uncensored, reasoning-focused model finetuned using ORPO and QLoRA on top of Qwen3-4B-abliterated-TIES.
1Element8 0.552,0.763,0.875,0.694,0.424,0.764,0.653
2Xiaolong 0.363,0.402,0.622,0.574,0.314,0.723,0.630
3
4Eva-Xiaolong-Heretic
5qx86-hi 0.548,0.749,0.870,0.686,0.426,0.764,0.651pip install mlx-lm1from mlx_lm import load, generate
2
3model, tokenizer = load("Qwen3-4B-Element8-Eva-Xiaolong-Heretic-qx86-hi-mlx")
4
5prompt = "hello"
6
7if tokenizer.chat_template is not None:
8 messages = [{"role": "user", "content": prompt}]
9 prompt = tokenizer.apply_chat_template(
10 messages, add_generation_prompt=True, return_dict=False,
11 )
12
13response = generate(model, tokenizer, prompt=prompt, verbose=True)