dealignai/Qwen3.5-VL-122B-A10B-UNCENSORED-JANG_2S

Add vMLX app banner and runtime note to model card

capabilities: stamp reasoning/tool parser + cache_type for vmlx CapabilityDetector Tier-1

fix: update Twitter/X handle to @dealignai

fix: add generation_config.json with eos_token_id to prevent thinking loops

fix: add eos_token_id to config.json

Upload README.md with huggingface_hub

Upload chat_template.jinja with huggingface_hub

Update README with full MMLU per-subject, HarmBench, branding

Add dealign logo

Add dealign mascot

Add files using upload-large-folder tool

initial commit