comfyui: fix in_channels to use shape[0] (latent input dim)

2026-09-21 20:31:39 -07:00
parent 5fe57f42fa
commit 43b8258061
2 changed files with 2 additions and 1 deletions
@@ -37,3 +37,4 @@ RUN chmod +x /opt/ComfyUI/comfyui-start.sh
 CMD ["/opt/ComfyUI/comfyui-start.sh"]
 EXPOSE 8188
 # v2: VAE 4D-to-5D latent expansion for video VAEs (qwen_image21)
+# v3: fix in_channels to use shape[0] (latent input dim) not shape[1] (hidden dim)
@@ -9,7 +9,7 @@ if "qwen_image21" not in code:
        "    if '{}txt_in.text_norm.weight'.format(key_prefix) in state_dict_keys:  # Qwen Image 2.1\n"
        "        dit_config = {}\n"
        '        dit_config["image_model"] = "qwen_image21"\n'
-        "        dit_config[\"in_channels\"] = state_dict['{}img_in.weight'.format(key_prefix)].shape[1]\n"
+        "        dit_config[\"in_channels\"] = state_dict['{}img_in.weight'.format(key_prefix)].shape[0]\n"
        "        dit_config[\"num_layers\"] = count_blocks(state_dict_keys, '{}transformer_blocks.'.format(key_prefix) + '{}.')\n"
        "        return dit_config\n"
    )