model
Qwen3.8-27B-OBLITERATED
A 27B MLX-format abliterated model — community curiosity, not a production workhorse
4.5Overall
Utility4
Onboarding7
Craft4
Niche fit5
Longevity3
Five dimensions scored independently; overall is a weighted average. Scores are only comparable within this same rubric.
Good for
Experimenters wanting uncensored local inference on Apple Silicon
Not for
Teams needing reliable, production-grade outputs or formal benchmark guarantees
Project description
text generation · mlx
Alternatives
Qwen2.5-32B-InstructDolphin-Mistral-24BLlama-3.1-8B-Instruct