<h2 id="qwen-for-minimax-h3-(qp)">Qwen for MiniMax H3 (QP)</h2><ul><li>This model has been quantized from a 100GB source QP training.</li><li>Only 2.2GB larger for NVFP4</li><li>This model HAS NOT been ablierated or uncensored by traditional means and is unlike any other model available currently</li></ul>The attached .json is a <a target="_blank" rel="ugc" href="https://civarchive.com/api/download/models/3332361?fileId=3218695">workflow</a> that speed my IT time up by 60%<hr />NOTE UPDATE YOUR COMFY &amp; <a target="_blank" rel="ugc" href="https://developer.nvidia.com/cuda-downloads">CUDA</a>Consider using --disable-dynamic-vram<ul><li>pytorch version: 2.13.0+cu132</li><li>xformers version: 0.0.35</li><li>ComfyUI version: 0.35.0</li><li> comfy-aimdo version: 0.5.5</li><li>comfy-kitchen version: 0.2.34</li></ul>

Qwen for MiniMax H3 (QP) **Fixing**

Files with SHA256: a6ce308a39865343e7e05fdc41f2a0ca6953ad3b0c0b8f17bf27e707e70facfc