Uncensored Qwen 3.8 27B MLX Build Released for Apple Silicon with Multiple Quantization Levels
alexcovo_eth · x · 2026-08-18
OrcaRouter released an uncensored MLX build of Qwen's 27B model tailored for Apple Silicon. The abliterated model removes safety refusals and is available in 2/4/6/8-bit quantization, running locally without CUDA or cloud dependencies. It retains vision-language, function-calling, and reasoning capabilities.
Related event: Uncensored Qwen3.8-27B Released for Apple Silicon via MLX(2 posts)→
More from Models
- Codex Still Disastrous at Interfaces, Corpus May Be Bootstrapped — wavefnx · 2026-08-18
- Dev: I'd never pick Anthropic via API, yet I keep coming back to Claude Code — bclavie · 2026-08-18
- The Powerful Chinese Model Experts Warned About—And Waited For—Is Here — Wired AI · 2026-08-18
- Google One vs. Code Assist: small business owner puzzles over the 1,500 daily request cap — innovaldragon · 2026-08-18
- Too Smart to Work With: Users Downgrade to Dumber Models for Readable Output — AashaySachdeva · 2026-08-18
- Muse Spark 1.2 release claims cost-intelligence Pareto frontier — zainhas · 2026-08-18