orcarouter / Qwen3.8-27B-Uncensored-MLX
Agents27B56.4K1.6KUpdated Oct 27, 2026
Downloads
56.4K
Likes
1.6K
Last updated
Oct 27, 2026
License
apache-2.0
An abliterated (refusal-removed) build of Qwen/Qwen3.8-27B — a 27B-parameter dense, hybrid-attention (Gated DeltaNet linear + full attention) native vision-language model with thinking control, tool-calling and an MTP head — quantized to MLX format for Apple Silicon. Four pre…
Tasks & library
- image text to text
- mlx
Tags
- mlx
- qwen3_5
- abliterated
- qwen3.8
- uncensored
- ai-red-team