MoringLabs/Qwen3.5-122B-A10B-MLX-3.7bit-VL

Update README.md

Update README.md

Upload config.json with huggingface_hub

Upload README.md with huggingface_hub

Fix tilde rendering as strikethrough

Revise Hardware Optimization: emphasize 64GB usability breakthrough

Update README with Hardware Optimization section and enriched metadata

Fix: original is not slow, just unusable at long context due to OOM/swap

Update README: clarify 96GB experience improvement

Add WikiText-2 perplexity evaluation

Show own perplexity scores only

Add WikiText-2 perplexity evaluation vs MINT 3bit

Upload Qwen3.5-122B-A10B mixed-precision 3.7bit MLX quant with vision support

initial commit