A text-focused model built on Qwen3.6's 27B architecture, published by AEON-7 with an 'uncensored' framing that signals reduced content filtering compared to base releases. This means it may respond more freely to edge-case or sensitive prompts, but that same looseness can mean less guardrail consistency. The NVFP4 quantization format suggests it's optimized for NVIDIA hardware efficiency at some cost to raw precision.