diff --git a/tutorials/image/qwen/qwen-image.mdx b/tutorials/image/qwen/qwen-image.mdx index 85af9fa4..698993ac 100644 --- a/tutorials/image/qwen/qwen-image.mdx +++ b/tutorials/image/qwen/qwen-image.mdx @@ -49,6 +49,7 @@ After updating ComfyUI, you can find the workflow file in the templates, or drag - Qwen-Image_bf16 (40.9 GB) - Qwen-Image_fp8 (20.4 GB) +- Distilled versions (non-official, requires only 15 steps) All models are available at [Huggingface](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/tree/main) and [Modelscope](https://modelscope.cn/models/Comfy-Org/Qwen-Image_ComfyUI/files) @@ -56,6 +57,16 @@ All models are available at [Huggingface](https://huggingface.co/Comfy-Org/Qwen- [qwen_image_fp8_e4m3fn.safetensors](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/diffusion_models/qwen_image_fp8_e4m3fn.safetensors) +The following models are unofficial distilled versions that require only 15 steps. +[Distilled Versions](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/tree/main/non_official/diffusion_models) +- [qwen_image_distill_full_bf16.safetensors](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/non_official/diffusion_models/qwen_image_distill_full_bf16.safetensors) 40.9 GB +- [qwen_image_distill_full_fp8.safetensors](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/non_official/diffusion_models/qwen_image_distill_full_fp8_e4m3fn.safetensors) 20.4 GB + + + - The original author of the distilled version recommends using 15 steps with cfg 1.0. +- According to tests, this distilled version also performs well at 10 steps with cfg 1.0. You can choose euler or res_multistep according to your desired image type. + + **Text Encoder** [qwen_2.5_vl_7b_fp8_scaled.safetensors](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors) diff --git a/zh-CN/tutorials/image/qwen/qwen-image.mdx b/zh-CN/tutorials/image/qwen/qwen-image.mdx index b6e45d95..89f3494f 100644 --- a/zh-CN/tutorials/image/qwen/qwen-image.mdx +++ b/zh-CN/tutorials/image/qwen/qwen-image.mdx @@ -46,6 +46,8 @@ import UpdateReminder from '/snippets/zh/tutorials/update-reminder.mdx' **ComfyUI 提供的版本** - Qwen-Image_bf16 (40.9 GB) - Qwen-Image_fp8 (20.4 GB) +- 蒸馏版本 (非官方，仅需 15 步) + 所有模型均可在 [Huggingface](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/tree/main) 或者 [魔搭](https://modelscope.cn/models/Comfy-Org/Qwen-Image_ComfyUI/files) 找到 @@ -53,6 +55,16 @@ import UpdateReminder from '/snippets/zh/tutorials/update-reminder.mdx' [qwen_image_fp8_e4m3fn.safetensors](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/diffusion_models/qwen_image_fp8_e4m3fn.safetensors) +下面的模型为非官方仅需 15 步的蒸馏版本 +[蒸馏版本](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/tree/main/non_official/diffusion_models) +- [qwen_image_distill_full_bf16.safetensors](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/non_official/diffusion_models/qwen_image_distill_full_bf16.safetensors) 40.9 GB +- [qwen_image_distill_full_fp8.safetensors](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/non_official/diffusion_models/qwen_image_distill_full_fp8_e4m3fn.safetensors) 20.4 GB + + + - 蒸馏版本原始作者建议在 15 步 cfg 1.0 +- 经测试该蒸馏版本在 10 步 cfg 1.0 下表现良好，根据你想要的图像类型选择 euler 或 res_multistep + + **Text Encoder** [qwen_2.5_vl_7b_fp8_scaled.safetensors](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors)