library_name: transformers
tags:
- Uncensored
- Abliterated
- Cubed Reasoning
- QwQ-32B
- reasoning
- thinking
- r1
- cot
- deepseek
- Qwen2.5
- Hermes
- DeepHermes
- DeepSeek
- DeepSeek-R1-Distill
- 128k context
- not-for-all-audiences
- merge
- mlx
- mlx-my-repo
base_model: DavidAU/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored
cs2764/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-mlx-6Bit-gs32
The Model cs2764/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-mlx-6Bit-gs32 was converted to MLX format from DavidAU/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored using mlx-lm version 0.26.2.
Quantization Details
This model was converted with the following quantization settings:
- Quantization Strategy: 6-bit quantization
- Group Size: 32
- Average bits per weight: 7.000
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("cs2764/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-mlx-6Bit-gs32")
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)