Apps & Tools
Qwen 3.8 27B Uncensored MLX
Uncensored, refusal-removed Qwen 3.8 27B build optimized for MLX and Apple Silicon.
Playable
Description
An abliterated (refusal-removed) build of Qwen 3.8 27B featuring hybrid Gated DeltaNet attention and 262K context. It is quantized to MLX format for high-performance inference on Apple Silicon, offered in 2, 4, 6, and 8-bit precisions with its vision-language capabilities preserved.
Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.